Education Trajectories: From Early Childhood to Early Adulthood … · 2016. 12. 9. · EDUCATION TRAJECTORIES: FROM EARLY CHILDHOOD TO EARLY ADULTHOOD IN INDIA 4 The authors Renu

November 2016

Renu Singh and Protap Mukherjee

Education Trajectories:From Early Childhood to Early Adulthood in India

Country Report

Education Trajectories: From Early Childhood to Early Adulthood in India

Renu Singh and Protap Mukherjee

© Young Lives 2016 ISBN 978-1-909403-82-6

A catalogue record for this publication is available from the British Library. All rights reserved. Reproduction, copy, transmission, or translation of any part of this publication may be made only under the following conditions:

• with the prior permission of the publisher; or

• with a licence from the Copyright Licensing Agency Ltd., 90 Tottenham Court Road, London W1P 9HE, UK, or from another national licensing agency; or

• under the terms set out below.

This publication is copyright, but may be reproduced by any method without fee for teaching or non-profit purposes, but not for resale. Formal permission is required for all such uses, but normally will be granted immediately. For copying in any other circumstances, or for re-use in other publications, or for translation or adaptation, prior written permission must be obtained from the publisher and a fee may be payable.

Young Lives, Oxford Department of International Development (ODID), University of Oxford,

Queen Elizabeth House, 3 Mansfield Road, Oxford OX1 3TB, UK

Tel: +44 (0)1865 281751 • Email: [email protected]

Core funded by

EDUCATION TRAJECTORIES: FROM EARLY CHILDHOOD TO EARLY ADULTHOOD IN INDIA

3

Contents The authors 4 Acknowledgements 4 Summary 5 11.. Introduction 7 22.. Background of the education system in India 7

2.1 Policy background 8 2.2 National statistics 12

33.. Methodology 19 3.1 Data 19 3.2 Tests in the Young Lives study 20 3.3 Sample 21 3.4. Main variables 22

44.. Educational outcomes: gaps over time 47 55.. Achievements and challenges in education 49 References 53 Appendices 55

Appendix A 55 Appendix B 55 Appendix C 56 Appendix D 57 Appendix E 58 Appendix F 59 Appendix G. Timeline of key educational initiatives and policies 60


4

The authors Renu Singh has over 25 years’ experience in teaching, teacher education, education policy

analysis and research, both in India and abroad. Trained as a Montessorian and special educator, her doctoral study was on the inclusion of marginalised children. Her special interests remain early childhood development, teacher education, inclusion and gender. She

has held a number of prestigious positions at NGOs, including Save the Children, and in university departments. She has also advised the Indian Government by serving on a variety of working groups, committees and boards. She is currently the Country Director at Young

Lives India and Visiting Professor at Jamia Millia Islamia University, New Delhi.

Protap Mukherjee is a Quantitative Research Associate at Young Lives India. He has an

MSc in Geography from Banaras Hindu University, Varanasi and an MPS (Master of Population Studies) from the International Institute for Population Sciences (IIPS), Mumbai.

He also has research experience at Jawaharlal Nehru University, New Delhi. Before joining Young Lives, he worked at IIPS, the National Population Stabilisation Fund, Microsoft Corporation India and Jawaharlal Nehru University.

Acknowledgements We are grateful to Santiago Cueto for providing the framework and useful comments on the

early draft of this paper, and to Alejandra Miranda for rendering his support to us during the writing of this paper.

About Young Lives

Young Lives is an international study of childhood poverty, following the lives of 12,000 children in 4 countries (Ethiopia, India, Peru and Vietnam) over 15 years. www.younglives.org.uk

Young Lives is funded by UK aid from the Department for International Development (DFID).

The views expressed are those of the author(s). They are not necessarily those of, or endorsed by, Young Lives, the University of Oxford, DFID or other funders.


5

Summary This report draws upon Young Lives longitudinal data gathered in Andhra Pradesh and

Telangana to trace the educational trajectories of two cohorts of children since 2002. From this data, it is clear that huge disparities exist in educational outcomes for children, based on

wealth index, gender, location and dynamic poverty status. Stratification of better-off children and boys into private low-fee charging schools adds further to the inequity.

Disparity begins in early childhood and we find that a large majority of children in urban areas

were attending private schools, with only 9 per cent of the wealthiest children enrolled in

public schools. Using poverty dynamics analysis we find that the percentage of children who were not enrolled in school at age 5 is 18 percentage points lower for those from chronically poor households, and that only 1.4 per cent of children from these households were enrolled

in private school, as against 58 per cent of those from least-poor households. These differences impact on learning outcomes at a young age: while children attending a private preschool achieved an average score in PPVT (Peabody Picture Vocabulary Test – an

instrument used to measure receptive vocabulary) of 33.3, public preschool children scored only 21.7. Children attending private preschools therefore appear to have an advantageous start compared to those attending public preschools (either because of characteristics of the

schools themselves, or the children’s background characteristics), something which then has long-term effects on their mathematics achievement at ages 8 and 12.

While enrolment among both the Younger Cohort (born in 2001-02) and Older Cohort (born

in 1994-95) at age 8 was close to universal (99.3 per cent and 98.3 per cent respectively), by

the age of 12 there was a larger number of the Younger Cohort enrolled in school (97.4 per cent) compared to the Older Cohort (90.8 per cent). In addition to this, we also observe a slow but steady increase in enrolment to private schooling, particularly for children from the

wealthiest families and socially advantaged groups of children: in 2009, 50.5 per cent of the Younger Cohort were enrolled in private schools, but only 24.7 per cent of the Older Cohort at the same age (8 years) in 2002. This is a trend that has been also observed across other

Indian states and internationally, raising potential concerns about the segregation of poorer children in public schools.

Concurrently, we also find that the number of children who are ‘overage’ (i.e. they are older

than expected for the grade they are in) is much higher among the Younger Cohort at age 8 and age 12 than the Older Cohort at the equivalent age, which is very surprising. Since 2010,

as a result of the Right to Education Act (RTE Act), there is a ‘no detention’ policy in schools which we would expect to have led to a decrease in repetition of grades. The increase in private school enrolment could be a possible reason for this continuation, since these

privately run schools, particularly in rural areas, remain unregulated.

Despite increasing enrolment, we observe a dip in learning levels for the 12-year-olds in

2013, as against 12-year-olds in 2006. Mathematics test scores remain lower for Younger Cohort children, raising issues related to learning outcomes and suggesting that the focus to

date has been on access rather than quality assurance.

At age 15, approximately 23 per cent of the Older Cohort children were no longer enrolled in

school. This highlighted both gender and poverty dimensions: while two out of five girls from chronically poor households were likely to not be enrolled in school at age 15, only one out of

ten girls were out of school among the least-poor households. In terms of cognitive


6

achievement scores for mathematics and reading at age 15, the pattern is similar to that observed for middle childhood, with relatively small differences by gender, but large differences seen by place of residence, maternal education and baseline wealth index.

Overall enrolment in private schools at age 15 is marginally higher (27.7 per cent) than observed at age 12 for the Older Cohort, but with very large differences according to wealth status, level of maternal education and caste. More boys are found to be enrolled in private

schools than girls across the wealth quintiles, with three out of four boys and more than half the girls (57 per cent) belonging to the consistently least-poor households enrolled in private schools at age 15. We also find that more girls had dropped out of school by age 15, among

chronically poor households (46.6 per cent), compared to a much smaller proportion (9.1 per cent) from the least-poor households.

By age 19, around 71 per cent of the Older Cohort had completed secondary education and

around 28 per cent of the sample had dropped out of school – the majority of whom were

girls, belonged to the poorest quintile, had mothers with low levels of education, and lived in rural areas. There are very large differences in dropout rates by age 19 among girls and boys belonging to households with differing dynamic poverty status: among chronically poor

households, more girls (46.6 per cent) than boys (33.3 per cent) were found to have dropped out of school, compared to the least-poor households where fewer girls (9.1 per cent) than boys (11.9 per cent) had dropped out. We also find that huge disparities exist among those in

full-time study at age 19, as well as in mathematics scores. These are associated with background characteristics such as level of maternal education, caste, and baseline wealth index, while gender plays a relatively small role. Only 37 per cent of children were in tertiary

education, and there was a clear gender bias in favour of boys.

This report highlights the increasing inequities and lack of opportunities afforded to children,

particularly girls, from chronically poor households. The low educational achievement for children from economically and socially disadvantaged backgrounds needs to be addressed

urgently to ensure that education becomes an equaliser rather than a source of increasing division. This would need consistent policies, programmes and public investment targeting educationally vulnerable children from an early age.

Some policy recommendations are:

• It is critical to focus on early years and ensure that preschool and primary teachers

are effectively trained to develop early literacy, so that children acquire skills and

retain interest in learning.

• Given the increase in private school enrolment of children from better-off households,

it is important that the Sustainable Development Goals’ focus on ‘quality education’ must become an inherent right for every child irrespective of gender, caste, ethnicity,

religion, socio-economic background, ability, or location.

• Since gender disparity in schooling increases as children grow older, safety nets and

social protection for the poorest households must be made available.

• There is a strong case for an autonomous department of standards and evaluation to

be set up and quality standards developed and implemented across all schools. A ‘common school system’ may be the best way forward to ensure that the education

system is able to promote equity and social justice.


7

1. Introduction Since independence in 1947, India has formulated policies and programmes to give impetus

to formal education from primary to tertiary level. As home to 19 per cent of the world’s child population, India boasts of the largest education system in the world, catering to

approximately 198.9 million students at elementary level (Grades I-VIII) and 59.6 million students at the secondary and senior secondary levels (IX-XII) (U-DISE 2015). However, despite considerable progress in enrolment at primary level, retention and smooth transition

of children from primary to secondary schools and beyond, along with achieving desirable learning levels, remain challenges.

This paper draws upon evidence gathered by Young Lives since 2002 on educational

trajectories of children in Andhra Pradesh and Telangana in southern India to analyse some

critical issues that need policy attention, such as disparity in educational outcomes related to gender, wealth status of household, location, and caste disadvantage. The paper comes at an opportune time, when India is poised to meet the goal of providing ‘inclusive, equitable

quality education’ and ‘lifelong learning opportunities’ as envisaged in Sustainable Development Goal 4. In order to do undertake the analysis, we take a dynamic view of the educational evolution of two cohorts of children from the two states that were born around

1994 (Older Cohort) and 2001 (Younger Cohort), examining them at different stages of development while undertaking cross-cohort comparisons of the two cohorts at the same age. We also draw upon other state and national research studies. Section 2 provides a

background of the education system in India and Section 3 outlines the Young Lives methodology. The subsequent sections then provide evidence related to pre-primary/early childhood, middle childhood/early adolescence, late adolescence, and early adulthood.

2. Background of the education system in India India is a vast country comprising 29 states and seven union territories with diverse socio-cultural contexts and widely varying geographical and climatic conditions. Under a federal

structure, the Central Government and the states share responsibilities for the planning and implementation of national development programmes. There are well-defined constitutional provisions and mechanisms for sharing of resources and responsibilities between the Central

Government and the states. The Constitution was amended in 1976 to change education from a state subject to a concurrent one, which implies that the responsibility for development of education is shared by the Central Government and state governments. Five-Year

National Development Plans guide the plans for education, with the key aims of the Twelfth Five-Year Plan (2012-17) being to reduce dropout rates at the elementary level to below 10 per cent, and raise the gross enrolment ratio (GER) at the secondary level to over 90 per

cent and to over 65 per cent at the higher secondary level.

There are four broad stages of school education in India – primary, upper primary, secondary

and higher secondary. The primary and upper primary stages constitute the elementary stage. The national system of education envisages a 10+2 pattern of school education; five years of

primary, three years of upper primary, two years of secondary, and two years of higher


8

secondary education. At the tertiary level undergraduation may be for three to four years duration, followed by Master’s degree of two years. The pre-primary stage, which is critical for laying the foundation for primary education, is currently not a part of the formal education

structure; while the primary and upper primary classes together are called ‘elementary classes’ and are covered under the Right to Free and Compulsory Education Act, 2010.

2.1 Policy background

Article 45 of the Indian Constitution states that the state shall endeavour to provide within ten

years of commencement of the constitution free and compulsory education to all children up to the age of fourteen years by 1960 (Government of India, 2007). However, since Article 45

remained a directive principle, this commitment was only realised when Article 21-A was inserted in Part III of the constitution after the 86th Amendment Act in 2002. It reads:

21-A Right to Education: The State shall provide free and compulsory education to all

children of the age of six to fourteen years in such manner as the State may, by law,

determine.

The Education Commission (1964-66), also known as the Kothari Commission,

recommended a common school system of public education as the basis of building up the National System of Education with a view to bringing the different social classes and groups together and thus promoting the emergence of an egalitarian and integrated society. This

was followed by the National Policy on Education (1968) which established a common structure of education throughout the country, while the National Policy of Education (1986) proposed the National Literacy Mission, which was launched in 1988, emphasising the

removal of disparities and equalisation of educational opportunities of disadvantaged groups such as Scheduled Castes, Scheduled Tribes and minorities. Building on this further was the Programme of Action (1992), which accorded an unqualified priority to Universalisation of

Elementary Education (UEE) and received a further fillip with the Jomtien Declaration in 1990, which addressed universalisation of access and promotion of equity as essential goals for basic education. This was followed by the District Primary Education Programme (DPEP)

in 42 districts nationwide, funded by the World Bank, which promoted decentralised management. Sarva Shiksha Abhiyan (SSA) was yet another centrally sponsored scheme which began in 2002, aiming for universalisation of elementary education with the goals of:

(a) all children in elementary school; (b) bridging gender and social gaps; (c) all children retained in elementary education; and (d) education of satisfactory quality.

In India, schools are broadly grouped into three categories: (1) public or government schools

run by various state education departments, which are government owned and controlled

and which do not charge fees; (2) private aided schools, which are schools managed by private bodies but which receive government funds to pay teachers’ salaries, which also do not charge fees; and (3) private unaided schools, which are privately managed, do not get

any aid from the government, and charge students fees. Private unaided schools frame their own admission rules and fee structure, and tuition fees vary from Rs. 30 to Rs. 3,000 per month depending on the location and services provided by the school. It is critical to mention

that private schools are not homogeneous and that is why one can find very low-fee charging schools in rural areas and urban slums of India and internationally competitive high-fee charging schools in urban areas, catering to rich students (Singh and Sarkar 2012).


9

A major development relating to education sector in India in the past few years has been the establishment of constitutional and legal underpinnings for achieving universal elementary education. A huge milestone in educational policy was the Right of Children to Free and

Compulsory Education Act 2009 (popularly called Right to Education Act or RTE Act) legislated in August 2009, which came into effect on 1 April 2010 (as a consequence of the 86th Amendment Act), whereby India made a legal commitment to provide free and

compulsory education for children aged 6-14 years. The main features of the Act included the proviso that states would identify out-of-school children between 6 and 14 years and provide them with an age-appropriate placement. Furthermore, norms for infrastructure and pupil-

teacher ratio at primary and upper primary level were specified and unaided private schools were asked to reserve 25 per cent of places for children from economically weaker backgrounds. Figure 1 presents the structure of the Indian education system.

Figure 1. Structure of education in India

One of the major omissions of RTE Act was that it did not address the critical foundation

years of children below the age of 6. This was in direct contradiction to the Supreme Court's

historic Unnikrishnan judgment in 1993, which gave all children up to 14 years of age a fundamental right to education. The court contended that the fundamental right to life (Article 21) of the Constitution should be read in 'harmonious construction' with the Directive in

Article 45 to provide free and compulsory education to children aged 0-14 years, including those below 6 years old. However, the 86th Constitutional Amendment Act, Article 21A, limited the fundamental right to education to 6-14 years and furthered this huge mistake by

not recognising the importance of the early years. The Twelfth Five-Year Plan by the Ministry of Human Resource Development (MHRD) also emphasised the need to maintain continuity from early childhood care and education (ECCE) to early primary grades to ensure a smooth

transition for children through integrated ‘early learning centres’ covering preschool and Grades 1 and 2. ECCE received a boost in 2013 with the formulation of the National Policy on ECCE, accompanied by a National Curriculum Framework and Quality Standards (MWCD

2013). India has the unique distinction of having the largest publicly sponsored integrated programme for children below 6 years old in the world: the Integrated Child Development Services (ICDS), with 1.3 million Anganwadis or ECCE centres established across most

areas in the country. The programme offers a package of six services, with ECCE or non-formal preschool education one of these, delivered through a single Anganwadi worker and a

Years /Grades I II III IV V VI VII VIII IX X XI XII XIII XIV XV XVI XVII XVIII XIX

Age 0 - 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

M.Phil Ph.D.

B.E. / B. Tech. M.Tech.

MBBS MD/MS

.

Early Childhood Education

Pre-Primary

Open University

Diploma Teacher Education

Poly Technics / ITI/Vocational Colleges

Elementary Education Secondary Education Higher Education

Primary School

Upper Primary School

Secondary School

Higher Secondary

School

Integrated Graduate Degree

Under Graduate CoursesPost Graduate

B.Ed.


10

helper. The recently restructured ICDS scheme also made a clear statement regarding converting all Anganwadi centres into ‘vibrant ECCE centres’.

Secondary education

Acknowledging that secondary education is a key link between basic education and

economic development, preparing adolescents to learn life skills and participate in the growth process, policymakers have in recent years turned their attention towards achieving universal

secondary education. The Central Advisory Board of Education Committee Report (2005) highlighted that ‘universal secondary education is a precondition for equitable social development, widening participation in India’s democratic functioning, building up of an

enlightened secular republic, and be globally competitive’ (14). To meet the challenge of dramatically improving equitable access to secondary education, as well as the quality of the education provided, the Government of India launched a centrally sponsored scheme for

secondary education, Rashtriya Madhyamik Shiksha Abhiyan (RMSA) (MHRD 2009). From a historical perspective, the significance of RMSA lies in the fact that for the first time, the Central Government has begun to support the states to increase access to, and improve the

quality of, secondary education, thereby bestowing on secondary education the attention it deserves (RMSA 2014). The scheme seeks to enhance enrolment in Grades 9 and 10 by providing a secondary school within a reasonable distance of every habitation, to improve the

quality of the education at secondary level by ensuring that all secondary schools conform to prescribed norms, to remove gender, socio-economic and disability barriers, and to achieve near-universal enrolment at secondary level education with the GER exceeding 90 per cent

by 2017 (the end of the Twelfth Five-Year Plan).

Vocational education

As shown in Figure 1, vocational education is covered under polytechnics, industrial training

institutes (ITIs) and secondary schools. Vocational education has received a strong fillip under the National Skill Development Policy (2009) which set the target of providing skills training of 500 million people by 2022, out of which about 50 million people are expected to

become skilled through programmes within the education sector. During 2013-14, 955,000 people were covered under vocational education and skilling programmes. The National Skill Qualification Framework (NSQF) was notified in December 2013 by the National Skill

Development Authority (NSDA), which was set up as an autonomous agency to coordinate and harmonise the skill development initiatives. In 2014, the centrally sponsored scheme of “Vocationalisation of Higher Secondary Education” was revised and renamed

“Vocationalisation of Secondary and Higher Secondary Education”. It aims to integrate general academic education, vocational education, vocational training, and higher education as a comprehensive system and introduces vocational education from Grade 9 onwards,

while offering performance-based incentives to government-aided and recognised private unaided schools to promote vocational education. Furthermore, to give impetus to skill development, the Ministry for Skill Development and Entrepreneurship was set up in 2014 to

help create an appropriate ecosystem that facilitates imparting employability skills to the growing workforce over the next few decades. A new National Policy for Skill Development and Entrepreneurship (2015) supersedes the 2009 policy, in order to meet the challenge of

improving skills at scale with speed, standard (quality) and sustainability.


11

Higher education

Higher education is on the ‘concurrent list’ in the Indian Constitution, meaning that it is a

shared responsibility between the Central Government and state governments. Higher education institutions consist of universities and affiliated colleges that can award degrees

to students after they complete 12 years of schooling. These institutions of higher learning are funded by the Central Government through the University Grants Commission (UGC), one of the statutory bodies, by the state governments, or are privately owned. India has a

three-tier degree structure with bachelor, master and research degrees. Universities also offer shorter programmes at certificate and diploma-level. The Twelfth Five-Year Plan seeks to increase gross enrolment ratio (GER) in higher education to 21 per cent by 2017 and 30

per cent by 2020. Rashtriya Uchchatar Shiksha Abhiyan (RUSA), a centrally sponsored scheme, was launched in 2013 and aims at providing strategic funding to eligible state higher educational institutions and improving quality by adopting a national quality

assurance framework. The statutory body for technical education is the All India Council for Technical Education (AICTE). AICTE decides on standards for technical courses conducted at institutions such as polytechnics, universities, Indian Institutes of Technology (IITs) and

regional engineering colleges. Technical and vocational education courses are offered at industrial training institutes.

Teacher education

The centrally sponsored scheme on teacher education that has been implemented through

state governments and union territories since 1987 (following NPE 1986) emphasises the significance and need for a decentralised system for the professional preparation and support of teachers. The scheme was modified in 2012, keeping in mind teacher

requirements and preparation for compliance with the RTE Act 2009. The scheme aims to help establish and nurture capacities in a range of institutions and programmes of teacher education. These include both pre-service and in-service interventions, along with evolution

of training plans and models, material development, research, monitoring and evaluation.

The National Council of Teacher Education (NCTE), established by an Act of Parliament in

1993, has the statutory powers to award recognition to teacher education institutions, and to develop teacher education programmes at diploma, bachelor and master level. While

diploma courses need not be run in colleges of higher learning, both BEd and MEd can only be conducted through colleges affiliated with universities. NCTE developed the National Curriculum Framework of Teacher Education (2009) which articulated the new vision of

teacher education as:

Understand the self and others, one’s beliefs, assumptions, emotions and aspirations;

develop the capacity for self-analysis, self-evaluation, adaptability, flexibility, creativity and innovation.

Develop habits and the capacity for self-directed learning, have time to think, reflect,

assimilate and articulate new ideas; be self-critical and to work collaboratively in groups.

Engage with subject content, examine disciplinary knowledge and social realities, relate

subject matter with the social milieu of learners and develop critical thinking.

Develop professional skills in pedagogy, observation, documentation, analysis and

interpretation, drama, craft, story-telling and reflective inquiry (2009: 24)


12

In 2015, keeping in view recommendations from the Justice Verma Commission Report

(2012) aimed at improving the quality of teacher education , a two-year BEd programme was initiated across all states for teaching upper primary and secondary classes, while a two-year

DEd programme prepares teachers for primary classes.

Private sector participation

Over the past decade there has been a growing participation of the private sector throughout

the education system, in the form of the establishment of schools and colleges across urban and rural areas at various levels. In 2013-14, 75.9 per cent of the 1.45 million schools imparting elementary education (Grades 1–8) in India were managed by various levels of

government, while the private-unaided school share remained at 17.4 per cent and private aided schools constituted 4.69 per cent. In Andhra Pradesh, government schools constituted 75.5 per cent of primary and upper primary schools, while private unaided schools

constituted 24 per cent and private aided only 3 per cent.

At the secondary level, private schools make up an even larger share of the total number of

schools. In 2013-14, the share of government schools nationally was close to 41.6 per cent, with private aided constituted 17.3 per cent and private unaided 38 per cent, while in Andhra

Pradesh the government, private aided and private unaided share was 52.5 per cent, 3.5 per cent and 42.2 per cent respectively.

At the higher education level too, the private sector plays a dominant role, with 267 out of

757 universities reported by MHRD 2014-15 to be privately owned. At college level, private unaided colleges have a 61 per cent of share of the total number in the country, while private

aided colleges occupy 15 per cent and government colleges 23.9 per cent. In Andhra Pradesh and Telangana, private unaided colleges occupy an even greater share than the national average, at 82.8 per cent and 82.3 per cent respectively (AISHE 2014-15).

2.2 National statistics

Table 1 shows the national data on educational institutions at various levels of education.

There has been an exponential increase in the number of schools as well as colleges in the

country, demonstrating the remarkable investment in institution building since 2000-01. Between 2000-14 the total number of primary schools (schools with only primary section) increased by 34.5 per cent (from 638,738 to 858,916 schools). The total number of schools

imparting only upper primary education has increased by 185.9 per cent (from 206,269 to 589,796), while the total number of schools imparting elementary education (schools with primary or upper primary sections, schools with primary and upper primary sections, and

secondary/higher secondary schools with primary and or upper primary section) has increased by 71.4 per cent (from 845,007 to 1,448,712) during the same period (UNESCO-UIS 2015).


13

Table 1. Educational institutions in India between 1950 and 2014

Year in '00s in absolute number

Primary Upper primary

Secondary Higher secondary

Colleges University

1950–51 2,097 136 NA 74 578 27

1960-61 3,304 497 NA 173 1,819 45

1970-71 4,084 906 NA 371 3,277 82

1980-81 4,945 1,186 NA 516 6,963 110

1990-91 5,609 1,515 NA 798 5,748 184

2000-01 6,387 2,063 877 384 10,152 254

2005-06 7,726 2,885 1,060 536 16,982 350

2006-07 7,849 3,056 1,122 574 19,812 371

2007-08 8,056 4,451 1,138 592 23,099 406

2008-09 8,091 4,764 1,221 642 27,882 440

2009-10 8,099 4,938 1,222 717 25,938 436

2010-11 8,272 5,350 1,312 720 32,974 621

2011-12 8,424 5,696 1,283 841 34,852 642

2012-13 8,538 5,778 2188 1,196 35,829 665

2013-14 8,589 5,897 2266 1,026 36,671 712

Source: U-DISE, NUEPA, Statistics of Higher and Technical Education, AISHE, MHRD 2014.

In the same period (2000-14), enrolment in primary and upper primary (i.e. elementary

classes) rose from 156.6 million in 2000-01 to 198.9 million in 2013-14 (NUEPA 2014 b), an overall increase of 42.3 million. The gender gap in enrolment also began to close during this period, with enrolment of girls increasing by 28.8 million (from 67.3 million to 96.1 million)

while that of boys increased by 13.5 million (from 89.3 million to 102.8 million).

Figure 2. Enrolment in elementary education (Classes I–VII) 2000-01 to 2013-14

Source: Statistics of School Education, 2007-08, MHRD, GoI; Educational Statistics at a Glance, 2011, MHRD, GoI; Statistics of School Education, 2010-11, MHRD, GoI; and U-DISE, NUEPA.

ASER (2015) reports that, between 2007 and 2013, total enrolment in primary schools in the

country peaked in 2011 at 137 million, while upper primary enrolment grew from 51 million to around 67 million. During this period, enrolment in government schools (Grades 1-8) declined

89.3

89.7

95.7

99.4

102.

1

102.

6

102.

7

102.

8

67.3

69.0

81.4

84.9

90.6

92.4

96.4

96.1

156.

6

158.

7 177.

1

184.

3

192.

7

195.

0

199.

1

198.

9

0

20

40

60

80

100

120

140

160

180

200

2000-01 2001-02 2003-04 2005-06 2007-08 2009-10 2011-12 2013-14

Mill

ion

Boys Girls Total


14

by about 11.7 million, from 133.7 million to 121 million. In contrast, the enrolment in private schools increased dramatically by 27 million, from 51 million to 78 million.

Availability of quality teachers is crucial to ensure that the goals of quality education are met.

Table 2 shows the efforts that have been made by state governments to recruit teachers in

the last decade, particularly at upper primary, secondary and senior secondary levels.

Table 2. Number of teachers in primary, upper primary, secondary schools and higher education in India 1950 to 2014 (in thousands)

Year Primary Upper primary Secondary Higher secondary

1950-51 538 86 NA 127

1960-61 742 345 NA 296

1970-71 1,060 638 NA 629

1980-81 1,363 851 NA 926

1990-91 1,616 1,073 NA 1,334

2000-01 1,896 1,326 1,006 756

2005-06 2,184 1,671 1,123 1,032

2006-07 2,323 1,717 1,173 1,075

2007-08 2,315 1,780 1,175 952

2008-09 2,229 1,899 1,194 1,024

2009-10 2,217 1,778 1,185 1,145

2010-11 2,099 1,887 1,247 1,261

2011-12 2,254 2,057 1,163 1,303

2012-13 2,656 2,427 944 1,799

2013-14 2,684 2,513 1,286 1,785

Source: Statistics for School Education, UDISE- NUEPA.

Between 1950 and 2013-14, the gross enrolment rate in elementary education increased

substantially, especially in upper primary schools.

Table 3. Gross enrolment rate for primary and upper primary from 1950 to 2013-14

Year Primary (I - V) 6-10 years old Upper primary (VI - VIII) 1 -13 years old

Boys Girls Total Boys Girls Total 1950-51 60.6 24.8 42.6 20.6 4.6 12.7

1960-61 82.6 41.4 62.4 33.2 11.3 22.5

1970-71 95.5 60.5 78.6 46.5 20.8 33.4

1980-81 95.8 64.1 80.5 54.3 28.6 41.9

1990-91 94.8 71.9 83.8 80.1 51.9 66.7

2000-01 104.9 85.9 95.7 66.7 49.9 58.6

2005-06 112.8 105.8 109.4 66.7 49.9 58.6

2006-07 114.6 108.0 111.4 75.2 66.4 71.0

2007-08 115.3 112.6 114.0 77.6 69.6 73.8

2008-09 114.7 114.0 114.3 81.5 74.4 78.1

2009-10 115.5 115.4 115.5 84.5 78.3 81.5

2010-11 115.4 116.7 116.0 87.7 83.1 85.5

2011-12 106.8 109.3 108.0 72.9 76.3 74.5

2012-13 104.8 107.2 106.0 80.6 84.6 82.5

2013-14 100.2 102.7 101.4 86.3 92.8 89.3

Source: Statistics of School Education, U-DISE, NUEPA.

The number of out-of-school children at the national level in the 6-14 age group was

estimated at 32 million in 2001 (Census 2001). Three studies commissioned by MHRD have revealed a huge reduction in this number, which came down to 13.45 million in 2005-06 and


15

6.04 million in 2013-14 (SRI-IMRB survey). SRI-IMRB survey (2014) found a higher proportion of girls (3.23 per cent) were out of school than boys (2.77 per cent), while more children from rural areas (3.13 per cent) were out of school than in urban areas (2.54 per

cent). The social composition of out-of-school children (SRI-IMRB 2006) indicates that 9.5 per cent of Scheduled Tribes (STs), 8.2 per cent of Scheduled Castes (SCs), and 5.6 per cent of Other Backward Class (OBC) and Other Castes (OC) children were out of school in

2006; these fell to 4.2 per cent, 3.2 per cent and 2.5 per cent for STs, SCs and OBC/OC respectively in 2014 (Figure 3). Despite this reduction in the number of out-of-school children across social groups, ST, SC and children with disabilities are still found to be more

vulnerable in comparison to OBC and OC children. In particular, national data shows that there are 21,39,943 children with disabilities in the 6-13 age group, out of whom 6,00,627 or 28.07 per cent remain out-of-school.

With regards to school dropout rates too, there are also differences according to children’s

place of residence. Overall, figures show that 2.5 per cent of children are found to have dropped-out of school in urban areas, in comparison to 3.1 per cent in rural areas. Gender differences in out-of-school population still persist, with 3.2 per cent of girls dropping out of

school by the age of 13 compared to 2.8 per cent of boys. As per 2014 estimates, the highest proportion of out-of-school children within the 6-13 age group is estimated to be in the East zone (4.0 per cent) and the lowest within the South zone (0.97 per cent).

Figure 3. Percentage of dropped-out children (aged 6-13 years) by caste and disability in India 2006-14

Source: Census 2001, reports of IMRB Survey 2005, 2009, 2014.

Between 2000 and 2014, enrolment in secondary/higher secondary education increased from

27.6 million to 59.6 million, while the GER rose by 24.9 percentage points to 76.6 per cent (for secondary education) and by 24.4 percentage points to 52.2 per cent for higher secondary.

Despite this progress, more than 16 million young adolescents of lower secondary school age were not enrolled in school in 2011 ( UNESCO-UIS 2015). Meanwhile the net enrolment rate (NER) was 45.6 per cent in secondary education (age 14-15) (U-DISE, NUEPA, 2015), and

30.4 per cent in higher secondary education (age 16-17) (U-DISE, NUEPA, 2015). At the higher education level, enrolment rates also increased, with the GER for higher education

8.2 9.5 5.6

38.1

6.0 5.6 2.7

34.1

3.2 4.2 2.5

28.1

0

10

20

30

40

50

Scheduled Caste Scheduled Tribe Other Backward Class / Others

Caste Disability

%

2006 2009 2014


16

doubling between 2004-05 and 2012-13, although much still needs to be done to increase the GER from a low 21.1 per cent in 2012-13.

Table 4. Gross enrolment rate for secondary and higher education for boys and girls from 2001 to 2013-14

Year Secondary (IX - X) 14-15 years

Higher secondary (XI - XII) 16-17 years

Higher education 18-23 years

Boys Girls Total Boys Girls Total Boys Girls Total

2001–02 NA NA NA NA NA NA 9.3 6.7 8.1

2002–03 NA NA NA NA NA NA 10.3 7.5 9

2003–04 NA NA NA NA NA NA 10.6 7.7 9.2

2004–05 57.4 45.3 51.7 30.8 24.5 27.8 11.6 8.2 10.0

2005–06 57.6 46.2 52.2 31.4 25.2 28.5 13.5 9.4 11.6

2006–07 58.6 47.4 53.5 31.5 26.1 28.9 14.5 10.0 12.4

2007–08 62.6 53.2 58.2 36.3 30.4 33.5 15.2 10.7 13.1

2008–09 64.2 55.0 59.8 37.0 31.2 34.3 15.8 11.4 13.7

2009-10 66.7 58.7 62.9 38.3 33.3 35.9 17.1 12.7 15.0

2010–11 69.0 60.9 65.2 42.2 36.1 39.3 20.8 17.9 19.4

2012–13 69.55 66.59 68.13 41.9 39.5 40.8 22.3 19.8 21.1

2013–14 76.8 76.5 76.6 52.8 51.6 52.2 NA NA NA

Source: Statistics for Higher and Technical Education, DISE and AISHE.

Excellent progress has been made in the gender parity of education, particularly at

elementary level where the Gender Parity Index (GPI) increased from 0.41 and 0.22 in 1950-51 (for primary and upper primary classes respectively) to 1.03 and 1.08 in 2013-14. This

progress is also seen at secondary level, where the GPI reached 1.00 and 0.98 in 2013-14 at secondary and higher secondary level respectively, although a gender gap remains in higher education, where the GPI was 0.89 in 2012-13.

Table 5. Gender Parity Index for primary, upper primary, secondary and higher education from 1950 to 2013

Year Primary (I - V)

Upper primary (VI - VIII)

Secondary ( IX - X)

Higher secondary (XI - XII)

Higher education

1950–51 0.41 0.22 NA NA NA

1960-61 0.50 0.34 NA NA NA

1970-71 0.63 0.45 NA NA NA

1980-81 0.67 0.53 NA NA NA

1990-91 0.75 0.61 NA NA NA

2000-01 0.82 0.75 NA NA NA

2005-06 0.94 0.88 0.8 0.8 0.69

2006-07 0.94 0.9 0.81 0.83 0.69

2007-08 0.98 0.91 0.85 0.84 0.7

2008-09 0.99 0.93 0.86 0.85 0.72

2009-10 1.00 0.94 0.88 0.87 0.74

2010-11 1.01 0.95 0.89 0.86 0.86

2011-12 1.01 0.99 0.93 0.92 0.88

2012–13 1.03 1.05 0.99 0.98 0.89

2013-14 1.03 1.08 1.00 0.98 NA

Source: Statistics of School Education, U-DISE, NUEPA, Statistics for Higher and Technical Education, and AISHE.


17

The next section provides education statistics related to undivided Andhra Pradesh, as

Young Lives research has been conducted there since 2002.1

2.2.1 Undivided Andhra Pradesh

Elementary level enrolment in undivided Andhra Pradesh has remained almost constant in

2011-13, with a GER of 96.74 at primary and 83.57 at upper primary level in 2013-14.

Figure 4. Enrolment in elementary education in undivided Andhra Pradesh.

Source: DISE 2013-14, NUEPA, New Delhi.

Meanwhile, undivided Andhra Pradesh had an NER of 78.31, 62.25 and 43.5 at primary,

upper primary and secondary level respectively, all of which were lower than the national average in both 2012-13 and 2013-14.

Table 6. NER at elementary and secondary level in Andhra Pradesh and India in 2012 and 2013

State 2012-13 2013-14

Primary Upper primary

Secondary Primary Upper primary

Secondary

Andhra Pradesh 81.78 60.12 41.31 78.31 62.25 43.56

India 90.78 64.24 41.90 88.08 70.20 45.6

Source: UDISE and SEMIS, NUEPA.

In Andhra Pradesh in 2013-14, 42.9 and 37.8 per cent of children enrolled at primary and

upper primary level were studying in private unaided schools, while 51.7 and 57.6 per cent were enrolled in government schools. A much smaller proportion (3.2 and 3.57 per cent

respectively) attended private aided schools (DISE 2013-14).

1 We do not have statistics for Telangana and Andhra Pradesh before 2014-15, therefore the statistics are provided for

undivided Andhra Pradesh.

7.44 7.24 7.22

3.81 3.85 3.87

0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00

2011-12 2012-13 2013-14

Mill

ions

Classes I - V Classes VI - VIII


18

2.2.1.1 Learning levels in Andhra Pradesh

The concern around low levels of learning highlighted earlier is also an issue in Andhra

Pradesh. The Annual Status of Educational Report (ASER) (2014) provides evidence for the learning levels of children studying in rural Andhra Pradesh in Grade 5 across government

and private schools. The study found that only 56 per cent of children in government schools and 57 per cent of private school children could read a Grade 2 level text accurately, highlighting the low level of reading proficiency among the children (see Figure 5).

Figure 5. Percentage of children in Standard V who can read Standard II level text in Andhra Pradesh, by school type from 2010 to 2014

Source: ASER 2014.

ASER findings similarly demonstrate low levels of numeracy among children in the state.

Figure 6 captures the declining trend in private school children’s ability to do division sums in Grade 5, showing a decrease of 10.5 percentage points since 2010, with 38.4 per cent of

children in private schools demonstrating computation skills in division as compared to 35 per cent of children in government schools.

Figure 6. Percentage of children in Standard V who can do division in Andhra Pradesh, by school type from 2010 to 2014

Source: ASER 2014.

57.0 57.0 59.9 54.5 55.9

66.7 67.4

58.6

66.8

57.1

0

10

20

30

40

50

60

70

80

2010 2011 2012 2013 2014

% o

f ch

ildre

n

Government Private

36.1 35.3 37 33.1 35.1

48.9 45.2

50.4 45.8

38.4

0

10

20

30

40

50

60

2010 2011 2012 2013 2014

% o

f ch

ildre

n

Government Private


19

As described in the previous section, Sarva Shiksha Abhiyan (SSA) launched in 2001,

followed by the Right to Education Act (RTE) in 2009, and has made huge strides towards ensuring access to schooling for even remote habitations. Furthermore, a lot of progress had

been made towards gender equity and enrolment of socially disadvantaged children. However, despite these positive steps, NER at upper primary level shows that a large number of children continue to drop out of school. In addition, although the transition rate

from primary to upper primary shows no gender disparity at 89.99 per cent and 89.50 per cent for girls and boys respectively, the transition rate from elementary to secondary school remains lower for girls, at 89.28 per cent, as compared to 93.76 per cent for boys (UDISE,

NUEPA 2014c). Examination of the elementary-to-secondary transition by caste shows very low transitions from elementary to secondary education among both girls and boys belonging to STs (85.47 per cent for boys and 83.25 per cent for girls) and Muslims (79.96 per cent for

boys and 79.68 per cent for girls), while high gender disparities exist among OBCs (94.18 per cent for boys and 88.49 per cent for girls) and SCs (92.18 per cent for boys and 87.67 per cent for girls). Those children on multiple axes of disadvantage are even less likely to remain

in education: Muslim and ST girls are least likely to transition successfully from elementary to secondary school and this gets worse at senior secondary level, where the overall transition rate is only 68.91 per cent in 2013-14. Clearly huge disparities exist between social groups

as children move from primary to secondary and senior secondary education, and it is important to move beyond focussing on access to smooth transitions and retention of the most disadvantaged children, with issues related to quality of education taking centre stage.

3. Methodology

3.1 Data

This study analysed data from the international longitudinal study on childhood poverty

conducted by Young Lives. Young Lives has followed 12,000 children over 15 years from India, Ethiopia, Vietnam and Peru. Young Lives follows two cohorts of children, the Younger

Cohort (born around 2001) and the Older Cohort (born around 1994). The study has information about the family and the child from four rounds of surveys carried out in 2002, 2006, 2009 and 2013. The Young Lives study in Andhra Pradesh has collected data on two

cohorts of children: 1,008 children born between January 1994 and June 1995, and 2,011 children born between January 2001 and June 2002. The children are located in twenty sites across seven districts in the former united Andhra Pradesh. The four rounds of household

survey provide rich information about the socio-economic background of the children's households, parental expectations/aspirations for the children, and also detailed child-specific data including cognitive and psycho-social tests and education related information.

Additionally, in 2010-11 school survey data was collected from a sub-sample of 953 Younger Cohort children who were 9 years old, studying in 249 schools located across nine districts of Andhra Pradesh. Stratified sampling was utilised to ensure representation of rural, urban,

public, and private aided and unaided schools (Singh and Sarkar 2015). In this study we present information about the four household surveys of both cohorts and the school survey.


20

Table 7. Young Lives design (approximate ages in years by round and cohort)

Round 1 (2002) Round 2 (2006) Round 3 (2009) Round 4 (2013)

Older Cohort 8 12 15 19

Younger Cohort 1 5 8 12

Source: Young Lives Study (2002, 2006, 2009, 2013). Note: The school survey was carried out in 2010, only for a subsample of the Younger Cohort.

3.2 Tests in the Young Lives study

In order to have information about learning achievement at different ages, Young Lives has

administered tests in mathematics, reading comprehension and receptive vocabulary in several rounds to both cohorts. This section explains the main characteristics of the tests which we refer to in this report.2

3.2.1 Tests administered to both cohorts

Peabody Picture Vocabulary Test (PPVT): This test measures children´s vocabulary. In the

test, the examiner shows four pictures and asks the child to choose the picture that represents the word read by the examiner. In the Young Lives study, we used the PPVT -III

adapted for use in Telugu. This test has 125 items and was administered to the Younger Cohort in Rounds 2, 3 and 4; and to the Older Cohort in Rounds 2 and 3. In the analysis in this paper, we have only used scores of children who answered PPVT in Telegu since it is

not possible to compare scores for those who answered in English.

3.2.2. Tests administered to the Younger Cohort

Maths test in Round 3 (2009-10): This test was divided in two sections. The first section had

nine items that measured basic operations, counting, knowledge of number, and number discrimination. The second section had 20 items that included addition, subtraction, multiplication and division with whole numbers.

Maths test in school survey (2010): The mathematics achievement test consists of two sections. The first section aimed at measuring basic quantitative and number concepts. It

included seven items on number identification, seriation and simple problem sums, with no time limit. The second section consists of 14 items which include two-digit and three-digit addition, subtraction, multiplication and division with a time limit of 60 minutes.

Maths test in Round 4 (2013): This test had 29 items aimed to assess skills in basic

mathematics operation with numbers (addition, subtraction, percentages and fractions), using both whole numbers and decimals, as well as skills in solving mathematics problems.

Reading test in Round 4 (2013): This test had two types of item. The first part had six items

where children had to link either a word or a sentence with one of the three images that best

represented that word/sentence, thereby measuring word and sentence comprehension. The second part had 18 items, aimed to measure reading comprehension, where children had to answer comprehension questions related to three texts.

2 For psychometric characteristics of tests in Round 3, see Cueto and Leon (2013).


21

3.2.3. Tests administered to the Older Cohort

Maths test in Round 3 (2009-10): This test was divided into two sections. The first section

had 20 items dealing with addition, subtraction, square roots, multiplication and division, using fractions and whole numbers. The second part had 10 items on mathematics problem-

solving, measurement, basic knowledge of geometry and data interpretation; these items were taken from the TIMMS3 and PISA4 publicly released set.

Reading in Round 3 (Cloze): In this test Older Cohort children were asked to read a sentence

and complete the missing words. This test measures knowledge in vocabulary and comprehension. The test had 24 items.

Maths test in Round 4 (2013): This test included 30 items that measured the child´s ability to

perform basic mathematics operations with two or more digits of decimals, fractions and whole numbers. It also measured the ability to use and apply math concepts in real-life situations.

Reading test in Round 4: This was a Telegu reading comprehension timed test which had

two parts. The first part measured child´s ability to read a word and simple text comprehension. The second part included items aimed to measure reading comprehension

by requiring children to retrieve information and make inferences based on three narrative texts. This test had 24 items in total.

Table 8. Summary of tests used in the report

Cohort R1 (2002) R2 (2006) R3 (2009) R4 (2013)

Older Cohort - PPVT PPVT, maths and reading (Cloze)

Maths and reading

Younger Cohort - PPVT PPVT and maths PPVT and maths

3.3 Sample

The original sample was randomly selected from 20 sentinel sites or clusters in undivided

Andhra Pradesh. Undivided Andhra Pradesh was divided into three regions — Coastal Andhra, Rayalaseema and Telangana – and administratively was divided into 23 districts,

which are further sub-divided into 1,125 mandals (a local term for an administrative division in Andhra Pradesh and Telangana). Since Young Lives is a study about childhood poverty, the sample mandals and villages in Andhra Pradesh were selected in 2001 using a semi-

purposive sampling strategy to identify regional variations and cover all disadvantaged groups (Kumra, 2008). As Table 9 shows, in Round 1 there were 2,011 children in the Younger Cohort (YC) and 1,008 in the Older Cohort (OC); in Round 4, 1,915 children remain

in the YC and 952 in the OC (attrition rates of 4.8 per cent and 5.6 per cent respectively). The attrition rate has been comparatively low due to the tracking exercise undertaken between rounds, and in this report, we use panel data related to 1,910 Younger Cohort children and

951 Older Cohort children in order to analyse trends of educational variables for the same group of children over time.5

3 The Trends in International Mathematics and Science Study (TIMSS) is a series of international assessments of the

mathematics and science knowledge of students around the world.

4 PISA, a collaborative effort among OECD member countries, assesses youth outcomes in three domains – reading literacy, mathematical literacy, and scientific literacy – through common international tests.

5 For cognitive achievement of children (as measured by different tests), sample sizes may be smaller as not all children appeared in the tests.


22

Table 9. Number of children by round and cohort

Round 1 (2002)

Round 2 (2006)

Round 3 (2009)

Round 4 (2013)

Panel

Younger Cohort 2,011 1,950 1,930 1,915 1,910

Older Cohort 1,008 995 977 952 951

3.4. Main variables

Given the background context summarised in the introduction, we focus on inequality as the main issue of analysis. Table 10 shows the main variables that are included in the inequality

analysis, along with samples for each category that have been used for the analysis.

1. Gender: Disparities between boys and girls have been widely analysed in India. In all

levels of education, there is a significant presence of gender disparity, particularly across wealth terciles and as children move to higher grades, as previous studies in

India using Young Lives data have also presented (Singh and Mukherjee 2015; Woodhead et al. 2009). Gender is a relevant variable for analysis nationally and internationally.

2. Place of residence: In India, disparities related to development indicators in general and education in particular are significantly prominent between rural and urban areas.6

Rural areas in Andhra Pradesh are also not receiving all the benefits of developmental processes that an urban centre can benefit from. Moreover, rural areas of Andhra Pradesh are also characterised by a larger Scheduled Caste/Scheduled Tribes

population who remain very vulnerable on all developmental indicators.

3. Maternal education: Using Young Lives data, several studies have shown that

maternal education is significantly associated with completion of secondary school education (Singh 2014; Singh and Mukherjee 2015). In this study, we divided maternal

education into the following categories: mothers with no formal education; completed primary; completed middle; and completed secondary and above levels of education.

4. Baseline wealth index: This index is a composite score based on assets at home,

access to public services and quality of the house infrastructure. We include in the

analysis the wealth index calculated from the first round of Young Lives survey in 2002 (when the Younger Cohort was around 1 and the Older around 8 years old), divided in quintiles, using the top and the bottom quintile.

5. Castes: Much research undertaken in India has examined the association between developmental indicators and castes. Previous studies show that among all the caste

groups, Scheduled Caste and Scheduled Tribes remain the most vulnerable, and the least likely to gain from the growth witnessed by the country in recent years. They are in the most disadvantaged position, followed by Backward Class (OBC), whereas

Other Castes remain the more privileged and socially and educationally advantaged caste (see Table A1 in Appendix A). In this report, we examine the association between caste and education at all levels.

6 According to Census of India, urban areas are defined as either places with a Municipality, Corporation or Cantonment or

Notified Town Area or all other places with the following criteria: (1) a minimum population of 5,000; (2) at least 75% of the male working population was non-agricultural; and (3) a density of population of at least 400 per square kilometre (1,000 per square mile). On the other hand, the village generally follows the limits of a revenue village that is recognised by the normal district administration. It may have one or more hamlets (censusofindia.gov.in).


23

6. Dynamic poverty status: Young Lives panel data enable us to examine poverty

mobility of households across time, including shifts of households across poverty levels. In this report, we have examined the educational indicators in response to the

dynamic poverty status of children from both cohorts. Only two categories of dynamic poverty status of the households have been considered: households which remained in the bottom tercile between Round 1 (2002) and Round 4 surveys (2013), termed

‘chronically poor’; and households which remained consistently in the top tercile between Round 1 and Round 4 surveys, termed ‘consistently least-poor’ households.

7. Gap analysis: This report is focused on measuring inequalities and gaps between

various groups of children based on gender, caste and other background variables.

Across various life stages such as early childhood, early and late adolescence we measure gaps by difference between the highest and lowest values in a group. This report also examines the gaps in educational indicators between consistently least-

poor and chronically poor children. The gaps between these two groups (i.e. difference between consistently least-poor and chronically poor children) have been presented in the tables to explore differences in educational outcomes for children

who were in chronically poor as well as least-poor households. To extend the analysis further, the gender dimensions of both chronically poor and consistently least-poor households are also presented in all the tables.

Table 10. Number of children by group and cohort Younger Cohort Older Cohort

Gender

Boys 1,026 469

Girls 884 482

Place of residence

Urban 457 225

Rural 1,453 726

Maternal education

No education 1,127 648

Primary 176 100

Middle 170 82

Secondary and above 427 98

Baseline wealth index (R1)

Top quintile 372 190

Bottom quintile 386 191

Caste

Scheduled Caste 351 204

Scheduled Tribe 283 105

Backward Class 890 439

Other Castes 386 203

Dynamic poverty status

Consistently least poor

– Boys 247 101

– Girls 187 99

Chronically poor

– Boys 187 102

– Girls 179 88

Source: Young Lives, both cohorts, R1 to R4 (2002-13).


24

3.4.1. Early childhood: preschool years

Over the past few decades, early childhood education has become a priority in policy-

making, mainly because it is known that education in this stage of life has long-term effects, even at secondary level (Singh and Mukherjee 2015). Education for All (EFA) included early

childhood education as the first goal, showing in the last EFA Global Monitoring Report that even though, on average, enrolment has increased there has been an uneven progress and quality still needs to be improved (UNESCO 2015).

In this section we capture preschool attendance of the Younger Cohort children who turned 5 years old in the Round 2 survey (for preschool attendance of the Older Cohort, see Appendix

B and Appendix E). Table 11 shows the enrolment in preschool of Young Lives children aged five in 2006. Data shows that the Younger Cohort had a higher enrolment in preschools as compared to the Older Cohort, and were more likely to attend private preschool,

disaggregated by variables mentioned earlier.

Table 11. Enrolment in preschool at age 5 (2006)

Younger Cohort Public (%) Private (%) Didn’t Attend (%)

Younger Cohort average 33.8 20.4 45.8

Gender

Boys 30.4 21.0 48.6

Girls 37.7 19.8 42.5

Gap -7.3 1.2 6.1

Place of residence

Urban 7.9 56.7 35.5

Rural 41.9 9.0 49.1

Gap -34.0 -47.7 -13.6

Maternal education

No education 41.4 7.5 51.1

Primary 33.5 15.3 51.1

Middle 31.2 27.7 41.2

Secondary and above 15.0 53.9 31.2

Gap 26.5 46.4 20.0


Top quintile 8.9 60.0 31.2

Bottom quintile 46.1 2.9 51.0

Gap -37.2 57.1 -19.9

Caste

Scheduled Caste 40.7 10.3 49.0

Scheduled Tribe 46.6 6.7 46.6

Backward Class 32.5 20.6 47.0

Other Caste` 21.0 39.4 39.6

Gap 25.7 32.7 9.4


Consistently least poor 9.2 58.1 32.7

– Boys 7.7 54.7 37.7

– Girls 11.2 62.6 26.2

Chronically poor 47.8 1.4 50.8

– Boys 46.0 2.1 51.9

– Girls 49.7 0.6 49.7

Gap (Least poor – poor) -38.6 56.7 -18.1

Older Cohort Average 41.0 18.9 40.1

Source: Young Lives, Younger Cohort, R2 (2006).


25

As expected, children in the top wealth quintile, those from urban areas, with mothers with

secondary and above education, and from Other Castes were more likely to attend private preschools. In India, relatively well-off families prefer to send their children to private

preschools as such preschools are considered to be of better quality than public preschools. Woodhead et al. (2009) highlighted that young children experience very early differentiation in their experiences and education opportunities, which are strongly shaped by factors such

as where they live and their household poverty levels.

In rural areas, 42 per cent of children are seen to be attending public preschools, while the

situation is reversed in urban areas, where 57 per cent of children are enrolled in private preschools. Based on wealth index, a huge disparity exists between children from the top

and bottom quintiles. While only 9 per cent of children from the top quintile were enrolled in public schools as against 60 per cent in private schools, this was reversed with only 3 per cent attending private schools and 46 per cent children attending public schools from the

bottom quintile.

The results obtained from analysis of chronically and least-poor households reveals that 58

per cent of children who belong to the consistently least-poor households were enrolled in private preschools, whilst only 1.4 per cent of the children from households who were

chronically poor were enrolled in private preschool. As expected, the percentage of children enrolled in public preschool from consistently non-poor households is much lower (at 9.2 per cent) than children from chronically poor households (at 47.8 per cent). Furthermore, the

percentage of children who were not enrolled in any preschool at the age of 5 is higher in the chronically poor households as compared to consistently least-poor households, with an 18 percentage point gap between children from these households.

The gender dimensions of this variable show some interesting results regarding preschool enrolment at age 5. In consistently least-poor households, the percentage enrolment in both

private and public preschools (62.6 and 11.2 respectively) is higher among girls than boys (54.7 and 7.7 respectively), but in chronically poor households while more girls attended public preschools, more boys were enrolled in private preschools. It is also interesting to note

that a higher percentage of boys than girls were not enrolled in preschools at age 5 for both consistent top and bottom wealth index households, although the difference is small.

Concerning achievement in early childhood, Table 12 shows the score in the Peabody Test

(PPVT) of receptive vocabulary.7 The results are presented as percentage of correct items

for those children who took the examination in Telegu language only for comparability. Children at age 5 had an average score of 22.

7 PPVT results for both Younger Cohort at age 5, 8 and 12 and Older Cohort at age 12 and 15 by different background variables

are presented in Appendix F.


26

Table 12. Score in PPVT at age 5 (2006)

Younger Cohort at age 5 (2006)

Average 22.1

Gender

Boys 22.3

Girls 21.9

Gap 0.4

Place of residence

Urban 29.7

Rural 19.9

Gap 9.8

Maternal education

No education 17.6

Primary 23.7

Middle 25.0

Secondary and above 32.1

Gap -14.6


Top quintile 33.6

Bottom quintile 18.4

Gap 15.2

Caste

Scheduled Caste 18.0

Scheduled Tribe 27.4

Backward Class 19.9

Other Caste 26.5

Gap - 9.4


Consistently least poor 31.1

– Boys 31.5

– Girls 31.4

Chronically poor 18.7

– Boys 19.4

– Girls 18.4

Gap (Least poor – poor) 12.4

Pre-school attendance

Private 33.3

Public 21.7

No preschool 17.1

Gap 16.2

Source: Young Lives, Younger Cohort, R2 (2006)

Children from more educated mothers, top quintile households and urban areas performed

better in comparison to their counterparts. There is a relatively small difference between boys and girls. The gap between Scheduled Tribes and Scheduled Caste is 9.3 percentage points

and surprisingly Scheduled Tribe children perform better than all groups at this age on PPVT. These results are relevant as the age of testing coincides with the end of preschool. Analysis of the PPVT scores by the types of preschool attended shows that children who attended

private preschools (33.3) scored better than children who attended public preschool (21.7) with a gap of 11.6. In this category, children with the lowest scores were those who did not attend any preschool (17.1).


27

These differences in PPVT score are further accentuated by the differentials between

children at age 5 from consistently least-poor households and those belonging to chronically poor households. On average, children from consistently least-poor households had an

average score of 33.1, whereas children from chronically poor households had an average score of 18.7.

3.4.2 Middle childhood: primary and upper primary school

Since children in India are expected to enter Grade 1 of primary school by age 6, in this

section we present information on the age at which children were enrolled in primary and upper primary school, using Round 3 and 4 data for the Younger Cohort and Round 1 and 2

for the Older Cohort. Table 13 shows almost universal enrolment patterns at ages 8 and 12 years for the Younger Cohort, at 99.1 and 97.2 per cent at age 8 and 12 respectively. This is higher than enrolment at either age for the Older Cohort (97.6 and 89.2 per cent respectively).

Table 13. Enrolment at ages 8 and 12 by cohort

Younger Cohort Older Cohort

Enrolment at age 8 (2009)




Average 99.1 97.2 97.6 89.2

Gender

Boys 99.3 97.4 98.3 90.8

Girls 98.8 96.9 96.9 87.6

Gap 0.6 0.4 1.4 3.3

Place of residence

Urban 99.8 98.7 97.8 95.1

Rural 98.8 96.7 97.5 87.3

Gap 1.0 2.0 0.3 7.8

Maternal education

No education 98.7 95.6 97.2 85.6

Primary 98.9 98.3 98.0 98.0

Middle 100.0 99.4 98.8 98.8

Secondary and above 99.8 99.8 99.0 98.0

Gap 1.3 4.1 1.8 12.3


Top quintile 99.7 99.5 98.4 96.3

Bottom quintile 97.7 94.8 96.3 78.5

Gap 2.1 4.6 2.1 17.8

Caste

Scheduled Caste 98.6 97.7 98.5 85.3

Scheduled Tribe 98.9 96.1 95.2 86.7

Backward Class 98.9 96.5 97.7 88.4

Other Caste 100.0 99.0 97.5 96.1

Gap 1.4 2.9 3.3 10.8


Consistently least poor 99.8 99.8 98.0 96.5

– Boys 100.0 99.6 99.0 97.0

– Girls 99.5 100.0 97.0 96.0

Chronically poor 97.0 95.4 95.8 78.4

– Boys 97.8 94.7 98.0 81.4

– Girls 96.1 96.1 93.2 75.0

Gap (Least poor – poor) 2.8 4.4 2.2 18.1

Source: Young Lives, both cohorts, R1 to R4 (2002-13)


28

Analysis of school enrolment data shows that many of the inequalities in enrolment which

were visible for the Older Cohort have reduced or disappeared entirely. Gender analysis of school enrolment shows that at age 12, the male-female gap for Younger Cohort is almost

negligible (0.4 percentage points), and considerably lower than that for the Older Cohort at the same age in 2006, when there was gender gap of 3.3 percentage points. The highest gap in enrolment was observed between poorest quintile and top (least poor) quintile households

(17.8 percentage points) among Older Cohort children at age 12, a gap which was still present, although considerably lower, for the Younger Cohort (4.6 percentage points).

Further analysis of enrolment at ages 8 and 12 by dynamic poverty status shows that the

enrolment remained at the same level (99.8) for consistently least poor Younger Cohort

children (suggesting there was no dropout during this period. For the same group of children among the Older Cohort, enrolment dropped by 1.5 percentage points between the ages of 8 and 12. Among the chronically poor children too, the dropout rate between age 8 and 12 is

much larger among the Older Cohort (as seen in Table 13) in comparison to that seen for this group in the Younger Cohort.

Since children are supposed to enter Grade 1 by age of 6, we use the education data from

the Young Lives household survey to calculate the proportion of children who are ‘overage’ at

each grade level. Even in primary, children could be behind their normative grade (i.e. overage) due to late entry, grade repetition, or temporarily abandoning school.8 Table 14 presents the percentage of children who are overage by age 8 and 12 for both Younger and

Older Cohorts, based on the premise that children should be 6 years old in Grade 1.

Children from more educated mothers, top quintile households and urban areas performed

better in comparison to their counterparts. There is a relatively small difference between boys and girls. The gap between Scheduled Tribes and Scheduled Caste is 9.3 percentage points and surprisingly Scheduled Tribe children perform better than all groups at this age on PPVT.

These results are relevant as the age of testing coincides with the end of preschool. Analysis of the PPVT scores by the types of preschool attended shows that children who attended private preschools (33.3) scored better than children who attended public preschool (21.7)

with a gap of 11.6. In this category, children with the lowest scores were those who did not attend any preschool (17.1).

These differences in PPVT score are further accentuated by the differentials between

children at age 5 from consistently least-poor households and those belonging to chronically

poor households. On average, children from consistently least-poor households had an average score of 33.1, whereas children from chronically poor households had an average score of 18.7.

3.4.2 Middle childhood: primary and upper primary school

Since children in India are expected to enter Grade 1 of primary school by age 6, in this

section we present information on the age at which children were enrolled in primary and

upper primary school, using Round 3 and 4 data for the Younger Cohort and Round 1 and 2 for the Older Cohort. Table 13 shows almost universal enrolment patterns at ages 8 and 12 years for the Younger Cohort, at 99.1 and 97.2 per cent at age 8 and 12 respectively. This is

higher than enrolment at either age for the Older Cohort (97.6 and 89.2 per cent respectively).

8 Officially, grade repetition is not allowed in Grade 1.


29

Table 13. Enrolment at ages 8 and 12 years by cohort






Average 99.1 97.2 97.6 89.2

Gender

Boys 99.3 97.4 98.3 90.8

Girls 98.8 96.9 96.9 87.6

Gap 0.6 0.4 1.4 3.3

Place of residence

Urban 99.8 98.7 97.8 95.1

Rural 98.8 96.7 97.5 87.3

Gap 1.0 2.0 0.3 7.8

Maternal education

No education 98.7 95.6 97.2 85.6

Primary 98.9 98.3 98.0 98.0

Middle 100.0 99.4 98.8 98.8


Gap 1.3 4.1 1.8 12.3


Top quintile 99.7 99.5 98.4 96.3


Gap 2.1 4.6 2.1 17.8

Caste



Backward Class 98.9 96.5 97.7 88.4

Other Caste 100.0 99.0 97.5 96.1

Gap 1.4 2.9 3.3 10.8



– Boys 100.0 99.6 99.0 97.0

– Girls 99.5 100.0 97.0 96.0


– Boys 97.8 94.7 98.0 81.4

– Girls 96.1 96.1 93.2 75.0


Source: Young Lives, both cohorts, R1 to R4 (2002-13)

Analysis of school enrolment data shows that many of the inequalities in enrolment which

were visible for the Older Cohort have reduced or disappeared entirely. Gender analysis of school enrolment shows that at age 12, the male-female gap for Younger Cohort is almost

negligible (0.4 percentage points), and considerably lower than that for the Older Cohort at the same age in 2006, when there was gender gap of 3.3 percentage points. The highest gap in enrolment was observed between poorest quintile and top (least poor) quintile households

(17.8 percentage points) among Older Cohort children at age 12, a gap which was still present, although considerably lower, for the Younger Cohort (4.6 percentage points).

Further analysis of enrolment at ages 8 and 12 by dynamic poverty status shows that the

enrolment remained at the same level (99.8) for consistently least poor Younger Cohort

children (suggesting there was no dropout during this period. For the same group of children among the Older Cohort, enrolment dropped by 1.5 percentage points between the ages of 8 and 12. Among the chronically poor children too, the dropout rate between age 8 and 12 is


30

much larger among the Older Cohort (as seen in Table 13) in comparison to that seen for this group in the Younger Cohort.

Since children are supposed to enter Grade 1 by age 6, we use the education data from the

Young Lives household survey to calculate the proportion of children who are ‘overage’ at

each grade level. Even in primary, children could be behind their normative grade (i.e. overage) due to late entry, grade repetition, or temporarily abandoning school.9 Table 14 presents the percentage of children who are overage by age 8 and 12 for both Younger and

Older Cohorts, based on the premise that children should be 6 years old in Grade 1.

Table 14. Overage at ages 8 and 12 by cohort


Overage at age 8 (2009)




Average 42.6 46.7 29.7 33.9

Gender

Boys 50.0 53.7 30.3 34.6

Girls 34.1 38.6 29.1 33.2

Gap 15.9 15.1 1.2 1.4

Place of residence

Urban 39.9 42.2 35.1 37.6

Rural 43.5 48.1 28.1 32.6

Gap 3.6 -5.9 7.1 4.9

Maternal education

No education 46.2 52.2 30.8 35.5

Primary 39.3 48.9 23.2 29.9

Middle 40.0 38.8 32.9 33.3


Gap 10.9 17.7 9.7 12.6


Top quintile 34.2 33.9 32.8 33.3


Gap 10.1 -20.5 -4.1 -12.0

Caste



Backward Class 40.1 44.2 25.2 28.9

Other Caste 43.6 43.5 32.2 34.5

Gap 9.4 15.8 14.6 18.4



– Boys 43.3 44.1 32.7 30.6

– Girls 29.4 29.9 21.2 21.1


– Boys 57.1 65.2 38.0 47.0

– Girls 36.2 48.6 39.8 45.5

Gap (Least poor – poor) -9.6 -19.1 -11.8 -20.4


Not surprisingly, for both Younger and Older Cohorts, the proportion of overage children

increases as they grow older, which may be attributed to the repetition of grades. Given that

9 Officially, grade repetition is not allowed in Grade 1.


31

the Right to Education Act (2009) stipulated a ‘no-detention policy’ from Grades 1-8, we would assume that there would be fewer overage children among the Younger Cohort (who were affected by the policy) than the Older Cohort. However, our findings do not substantiate this –

rather they suggest the opposite. Table 14 shows that 46.7 per cent of children in the Younger Cohort are overage at age 12 in 2013 as compared to 33.9 per cent of Older Cohort children in 2006. Cross-cohort comparison also shows that many more rural children and boys are

overage among the Younger Cohort at ages 8 and 12, as compared to the Older Cohort. While the gender gap in the proportion of overage children is 16 percentage points for the Younger Cohort in 2010, it was only 1.2 percentage points for the Older Cohort in 2006.

In addition, for the Younger Cohort at age 12 there are large inequalities in the proportion of

overage children linked with wealth, maternal education and caste. The results for Younger Cohort children also reveal that the percentage of overage children was higher in chronically poor households (57.1 per cent) than consistently least-poor households (38 per cent) in

2013, with a gap of 19.1 percentage points. It is also important to note that the percentage of overage children at age 12 in 2006 (Older Cohort children) was lower than percentage overage in 2013, although the percentage gap between consistently least poor and

chronically poor grew marginally during the same period.

For children belonging to both consistently least poor and chronically poor households, more

boys are found to be overage at ages 8 and 12 across both cohorts, with particularly high numbers among the Younger Cohort.

A deeper analysis of school progression of Young Lives Younger Cohort children by years and grades may be helpful to understand the overage phenomenon better (see Table 15).

Table 15. Number of children attending school across years and grades (Younger Cohort)

Grade 2004 (3 Yrs)

2005 (4 Yrs)

2006 (5 Yrs)

2007 (6 Yrs)

2008 (7 Yrs)

2009 (8 Yrs)

2010 (9 Yrs)

2011 (10 Yrs)

2012 (11 Yrs)

2013 (12 Yrs)

Pre-primary 375 994 1201 589 249 107

First grade primary 3 39 385 849 495 231 66 18 8 2

Second grade 3 30 354 775 469 217 71 19 8

Third grade 4 26 332 749 525 214 77 22

Fourth grade 3 23 313 736 517 219 74

Fifth grade 3 22 300 729 506 228

Six grade 3 22 294 718 499

Seventh grade 1 22 291 710

Eighth grade 1 21 285

Ninth grade 1 22

Tenth grade 1

% of children in schools

19.8 54.2 84.8 95.3 98.3 99.2 97.7 97.7 97.4 96.9

Source: Young Lives, Younger Cohort, R1 to R4 (2002-13).

Drawing upon the educational history of the Younger Cohort captured in multiple survey

rounds, Table 15 reveals that enrolment in preschools started at age 3, with a large number of children attending more than one year of preschool. At age 6, 589 children remained

enrolled in pre-primary classes, and interestingly even at age 8, 107 children were found enrolled in preschool and 231 in Grade 1 – clearly all overage for their respective classes. Examining the educational history data for Younger Cohort children also allows us to

understand patterns of enrolment and dropout – for example we can see that enrolment


32

peaked at 97.7 per cent at age 10 (2011) and then decreased marginally to 96.9 per cent enrolment in 2013, highlighting that a small number of children had dropped out of school by age 12. From this data we can also observe 23 children who are definitely underage,

studying in secondary classes (Grades 9 and 10) at age 12.

Since enrolment in private schools is one of the possible causes that is attributed to grade

repetition (Galab et al. 2014), Table 16 analyses enrolment in private schools for both cohorts at age 8 and 12. The analysis clearly demonstrates that more children among the

Younger Cohort were enrolled in private schools as compared to the Older Cohort at both ages. In particular, there was a huge influx of Younger Cohort children (age 8) to private schools in 2009 (44 per cent), when compared to the Older Cohort at the same age in 2002

(23 per cent) (Galab et al. 2010).Although private school enrolment for the Younger Cohort decreases by 2.6 percentage points between 2009 and 2013, it remains almost 16 percentage points higher than that for the Older Cohort in 2006.

Table 16. Enrolment in private school at age 8 and 12 for Younger and Older Cohort



Older Cohort at age 8 (2002)


Average 43.9 40.5 23.6 24.3

Gender

Boys 50.5 46.6 24.7 27.5

Girls 36.3 33.5 22.4 21.2

Gap 14.2 13.1 2.3 6.3

Place of residence

Urban 81.2 72.2 67.6 63.7

Rural 32.2 28.8 9.9 12.4

Gap 49.0 43.4 57.7 51.3

Maternal education

No education 28.4 23.9 12.3 13.7

Primary 42.6 34.5 32.0 31.0

Middle 57.6 49.3 43.9 47.6


Gap 51.2 53.8 61.2 55.7


Top quintile 86.0 80.0 72.1 70.5


Gap 70.2 68.4 68.4 64.7

Caste`



Backward Class 43.6 41.9 20.3 21.6

Other Castes 72.5 67.7 50.2 50.2

Gap 48.9 49.6 39.9 40.9



– Boys 87.4 82.2 69.3 72.3

– Girls 84.0 81.3 62.6 58.6


– Boys 20.3 8.8 2.9 5.9

– Girls 7.3 5.0 5.7 1.1




33

While the total enrolment rate has increased by only 1.6 per cent for the Younger Cohort

compared to the Older Cohort, the proportion enrolled in private schools has doubled between the two cohorts. Given that private schools charge fees, it is no surprise that there

are gaps in private school enrolment by caste, maternal education, wealth, and place of residence. However, such gaps do lead to serious equity concerns, with increased enrolment far from evenly distributed and gender-based school choices more prevalent (Woodhead et

al. 2011). This is evident by the fact that a 74.9 percentage points gap existed in private enrolment at age 12 in 2013 between children from households that are consistently least poor and chronically poor , an increase from the 61.8 per cent gap which was present in

2006 between the same groups in the Older Cohort. Boys were more likely than girls to be enrolled in private schools at age 12 in both the top and bottom wealth terciles in 2013, although the gap between boys’ enrolment between chronically poor and least-poor

households was 73.4 percentage points.

It is important to highlight that the majority of private unaided schools carry the ‘English

medium’ tag and many parents chose these schools for their children since they believed they were of good quality. Language of instruction at school is a relevant issue as it

influences children’s ability to grasp concepts that are taught. Almost 85 per cent of the Young Lives sample had Telegu as their mother tongue. Table 17 shows 64 per cent of children whose mother tongue was Telegu reported attending a school where the medium of

instruction remained Telegu, while 36 per cent attended a school with English as the medium of instruction at the age of 12. On the other hand, among children with Urdu as their mother tongue, 68.5 per cent attended a school where the medium of instruction was English, while

the remaining Urdu-speaking children attended schools with Telegu as the medium of instruction.

Table 17. Language of instruction during primary for the Younger Cohort at age 12, in percentage (2013)

Language at home Language of instruction at school

Telegu (%) English (%) Total (N)

Telegu 63.6 36.4 1,512

Urdu 31.5 68.5 92

Odiya 70.6 29.4 17

Kannada 81.8 18.2 66

Others* 81.0 19.0 121

Note: *Others include children speaking Hindi, Marathi, Tamil and Tribal languages at home.

As mentioned above, analysis has revealed changing patterns between student background

and the type of school they attend. Woodhead et al. (2013) show that although private schooling is predominantly urban, expansion has been most rapid in rural areas, with private

school enrolment trebling between 2002 and 2009. The most financially and socially advantaged children (categorised as ‘Other Castes’) constitute the largest group accessing private schools, with a 25 per cent increase and making up 70 per cent of all private school

students. Enrolment of children from Backward Classes is also on the rise and has doubled, albeit from a lower starting point. Private schools charge fees and thus tend to have a student population with families that have more economic resources. Singh and Bangay

(2014) found that across all quartiles and for both Younger and Older Cohorts studying in private schools, educational expenditure remained significantly higher. Within the private school sector, the richest quartile was observed to spend 1.7 times more at primary and


34

twice as much at secondary level when compared to the poorest quartile. This highlights the variance in fees charged by schools within the private school sector. Furthermore, among members of the richest quartile who send their children to public schools, spending was

almost 13 times the amount the poorest quartile spent on education expenses at primary level.10 Considering that public schools do not charge fees, the difference can be accounted for by the money spent by better-off families on after-school tuition.

Differences are also found regarding the infrastructure across private and public schools.

Table 18 presents information about school infrastructure in public and private schools, and in general, we can see that public schools have fewer or poorer facilities. In a system that promotes equity, public schools should have higher resources as they cater to more children.

Table 18. Percentage of students attending schools with selected characteristics (2011)

School Infrastructure Private schools (%) Public schools (%)

Drinking water 95.9 68.0

Toilets 83.3 64.3

Separate staff room 61.1 20.0

Library 39.0 3.5

Playgrounds 69.0 76.9

Covered space for assembly 13.5 0.7

Source: Young Lives school survey (2011)

Table 19 highlights characteristics of teachers, drawing upon analysis from the school survey conducted in 250 schools (118 private unaided and 109 public schools). We can see that

public school teachers had more years of teaching experience and a higher number had professional teaching qualifications compared to teachers in private schools.

Table 19. Percentage of mathematics teachers with selected characteristics (2011)

Teachers’ characteristics Private schools (%) Public schools (%)

Average age 28.3 33.0

Attended university 83.6 79.7

Years of experience 4.6 7.7

Have B.Ed /Dip.Ed. /M.Ed 57.2 82.3

Have Master’s degree and above 6.3 0.5

In-service training during the last academic year 4.4 59.6

Lecturing as a teaching method 47.2 53.9

Questioning individual students 10.8 6.8

Almost every exercise/piece checked 81.9 40.5

Source: Singh and Sarkar 2012.

However, classroom observations carried out during the primary school survey in 2010 found

that lecturing seemed to be the norm in close to half of both the private and public schools observed. The fact that around 80 per cent of teachers in private schools were found to have

regularly corrected almost every exercise given to children, as against only 40.5 per cent of teachers in public schools, suggests that a wide gap exists in terms of providing vital feedback that children need to enhance learning (Singh and Sarkar 2012).

10 Educational expenses in government schools at primary level consist primarily of after-school tuition fees, as government

schools.


35

Ultimately, it is important to understand the academic achievement of children studying

across public and private schools, given that good quality teaching as perceived by the parents appears to be the most compelling reason for selecting a type of school (Singh and

Sarkar 2012). Table 20 presents information on the mathematical abilities of Young Lives children across both cohorts.11 Despite many more children being enrolled in private schools among the Younger Cohort, the average mathematics score for the Younger Cohort is lower

by 12.4 percentage points as compared to scores of the Older Cohort at the same age. Whilst gaps between girls and boys are negligible, the gaps by caste, maternal education, and wealth are larger, with Other Castes and top wealth quintile children performing better

than the other children.12

Table 20. Achievement in mathematics at age 12 (2013)



Differences in cross-cohorts'

achievement scores

Average 52.8 65.3 -12.4

Gender

Boys 52.9 66.5 -13.6

Girls 52.8 64.1 -11.3

Gap 0.1 2.3

Place of residence

Urban 59.0 71.3 -12.3

Rural 50.9 63.4 -12.5

Gap 8.1 7.9

Maternal education

No education 46.1 60.8 -14.6

Primary 54.9 73.7 -18.7

Middle 59.4 72.8 -13.4

Secondary and above 67.4 83.0 -15.6

Gap -21.3 -22.2


Top quintile 65.1 76.0 -10.8

Bottom quintile 42.4 60.2 -17.8

Gap 22.7 15.8

Caste

Scheduled Caste 48.5 58.0 -9.5

Scheduled tribe 46.6 73.0 -26.4

Backward Class 53.0 63.2 -10.2

Other Caste 61.0 73.1 -12.1

Gap -12.4 -15.1



– Boys 65.2 78.5 -13.4

– Girls 67.6 74.1 -6.5

Chronically poor 39.7 59.5 -19.8

– Boys 39.6 58.8 -19.3

– Girls 39.9 60.2 -20.4

Gap (Least poor – poor) 26.5 16.8

Note: The score is based on three common items in the mathematics test. Source: Young Lives study (2006, 2013).

11 Some findings related to mathematics achievement scores are in Appendix C, Figures C1 to C3, and Table C1.

12 The statistical association between mathematics scores at age 12 (Younger Cohort) and wealth is shown in Appendix D.


36

The analysis of achievement in mathematics by dynamic poverty status shows that children

from consistently least-poor households scored an average of 66.2 per cent at age 12 in 2013, which was 26.5 percentage points higher than the average scores of children from

chronically poor households (39.7 per cent) among the Younger Cohort.

There are no substantial gender differentials in mathematics achievement scores observed

among children belonging to consistently least-poor households, although girls perform marginally better than boys among chronically poor households in the Younger Cohort at age

12, and a 2.6 percentage point gender gap was observed in favour of girls among chronically poor Older Cohort children in 2006.

Table 21 shows the percentage of correct answers of three common items in the maths test

given to both cohorts at age 12, which are intended to be aimed specifically at the abilities of

children at this age. It is clear that the learning levels of children have declined among 12 year olds in 2013 as compared to 12 year olds in 2006. This is a matter of policy concern, since despite progress towards universalisation of school enrolment at the elementary level,

there does not seem to be enough emphasis on learning outcomes.

Table 21. Percentage of correct answers (common items of mathematics)

Younger Cohort (2013) Older Cohort (2006) Item

59.0% 74.1% Which of these is equal to 342?

75.2% 83.1% Which of these is name for 9740?

24.4% 38.6% A piece of rope 204 cm long is cut into 4 equal pieces. Which of these gives the length of each piece in centimetres?

Source: Young Lives study (2006, 2013)

3.4.2 Late adolescence: secondary school

Looking at enrolment at age 15 (Table 22), we find that on an average 77.7 per cent of Older

Cohort children were enrolled in schools (i.e. close to 22 per cent had already dropped out of school by this stage).


37

Table 22. Enrolment at age 15

Older Cohort at age 15 years

(2009) (%)

Average 77.7

Gender

Boys 81.0

Girls 74.5

Gap 6.5

Place of residence

Urban 87.1

Rural 74.8

Gap 12.3

Maternal education

No education 71.9

Primary 90.0

Middle 92.7


Gap 23.0


Top quintile 92.6


Gap 25.6

Caste

Schedule Caste 76.5


Backward Class 74.9

Other Caste 85.7

Gap 10.8



– Boys 94.1

– Girls 89.9


– Boys 73.5

– Girls 63.6


Source: Young Lives, Older Cohort, Round 3, 2009.

Table 22 also reveals that at age 15, there was 6.5 percentage point gender gap in

enrolment in favour of boys. As expected, percentage enrolment at age 15 is higher in urban areas than rural areas and with increasing mothers’ educational levels, the enrolment of

children also increases. Children from the Other Castes category and from top wealth quintile households show higher enrolment than children from Backward Classes and from the poorest quintile.

The analysis of enrolment at age 15 by dynamic poverty status of children shows that there

was a substantial enrolment gap of 23 percentage points between consistently least poor children (92 per cent) and chronically poor children (69 per cent). Gender gaps in enrolment in favour of boys were visible in both consistently least poor and chronically poor children, but

we find that the gap was much larger among children who remain chronically poor (9.9 percentage points) over time. Thus we find that, while two out of five girls from chronically


38

poor households were likely to not be enrolled in school at age 15, only one out of ten girls were out of school among the least-poor households. This highlights the gender differences within the relationship between poverty and education.

Children are expected to move on to secondary schools after completing Grade 8 around the

age of 15. Table 23 presents information on the percentage of Older Cohort children who are overage by age 15. The proportion who are overage at age 15 is higher than observed at age 12 (see Table 14), although the patterns are quite similar to those presented before.

Relatively little difference is found by gender and urban/rural location as compared to gaps across other variables such as maternal education, castes and wealth. The difference is particularly large by caste groups, and we find that 48 per cent of Scheduled Tribe and 48

per cent of children from the bottom wealth quintile are overage.

Table 23. Overage at age 15

Older Cohort at age 15 years

(2009) (%)

Average 36.0

Gender

Boys 36.8

Girls 35.1

Gap 1.7

Place of residence

Urban 38.3

Rural 35.2

Gap 3.1

Maternal education

No education 37.8

Primary 30.0

Middle 36.8


Gap 10.9


Top quintile 35.2


Gap -13.2

Caste

Schedule Caste 41.7


Backward Class 30.7

Other Caste 35.1

Gap 18.1



– Boys 33.7

– Girls 21.3


– Boys 50.7

– Girls 48.2

Gap (Least poor – poor) -21.9



39

The highest percentage of overage children at age 15 is observed among children belonging

to chronically poor households (49.6 per cent), 21.9 percentage points higher than the percentage of overage children from consistently least-poor households. Interestingly, there

are marginal gender differences (2.5 percentage points) in the percentage of overage children from chronically poor households, whereas a 12.4 percentage points gender gap is visible among children from consistently least-poor households, where a third of the boys are

overage, compared to one out of five girls.

Table 24 shows that overall enrolment in private schools at age 15 is marginally higher (4.3

per cent) than observed at age 12 (see Table 16), but with very large differences found by top and bottom wealth quintile, maternal education and caste. More boys are found to be

enrolled in private schools than girls across all quintiles, with 75 per cent of boys and 57 per cent of girls in consistently least-poor households enrolled in private schools at age 15.

Table 24. Enrolment in private school at age 15

Older Cohort at age 15

(2009) (%)

Average 27.7

Gender

Boys 32.0

Girls 23.4

Gap 8.6

Place of residence

Urban 57.3

Rural 18.5

Gap 38.8

Maternal education

No education 17.6

Primary 36.0

Middle 52.4


Gap 50.8


Top quintile 68.4

Bottom quintile 9.4

Gap 59.0

Caste

Scheduled Caste 12.3


Backward Class 26.0

Other Caste 52.7

Gap 40.4



– Boys 75.2

– Girls 56.6


– Boys 7.8

– Girls 3.4




40

The largest gap in terms of enrolment in private school at age 15 is observed between

children from consistently least-poor households (66 per cent) and those from chronically poor households (5.8 per cent). Only 8 per cent of boys and 3 per cent of girls from

chronically poor households were enrolled in private schools. As we have already seen from the analysis of school type enrolment at younger ages, stratification of children into private and public schools based on household poverty, gender and caste is apparent among the

Older Cohort at age 15.

Table 25 shows that by age 15, 67.8 per cent of the sample had finished elementary

education, 9.9 per cent were still in the process of completing elementary school and 22.3 per cent had dropped out of school, with more dropping out in rural areas (25.2 per cent).

Similar patterns to those previously discussed are apparent in the proportions of children completing elementary education, with children with more educated mothers, from other castes and top quintile more likely to complete elementary schooling on age.

Table 25. Highest level of education at age 15 (Older Cohort)

Completed elementary on

age (%)

Completed elementary overage (%)

Still in elementary (%)

Dropped-out (%)

Undivided Andhra Pradesh 49.7 18.1 9.9 22.3

Gender

Boys 51.2 18.3 11.5 19.0

Girls 48.3 17.8 8.3 25.5

Gap 2.8 0.5 3.2 -6.5

Place of residence

Urban 53.8 21.8 11.6 12.9

Rural 48.5 16.9 9.4 25.2

Gap 5.3 4.8 2.2 -12.3

Maternal education

No education 44.8 16.7 10.5 28.1

Primary 63.0 19.0 8.0 10.0

Middle 58.5 25.6 8.5 7.3


Gap -24.6 -1.7 3.4 23.0


Top quintile 60.0 22.6 10.0 7.4


Gap 25.5 3.8 -3.6 -25.6

Caste


Scheduled tribe 39.1 20.0 17.1 23.8

Backward Class 51.9 15.0 8.0 25.1

Other Caste 55.7 22.7 7.4 14.3

Gap -11.1 -3.5 5.4 9.2



– Boys 62.4 22.8 8.9 5.9

– Girls 70.7 14.1 5.1 10.1


– Boys 36.3 17.7 19.6 26.5

– Girls 33.0 19.3 11.4 36.4

Gap (Least poor - poor) 31.8 0.1 -8.8 -23.1



41

The largest gap in terms of completion of elementary schooling on age was observed

between children from consistently least-poor households (66.5 per cent) in comparison to children from chronically poor households (34.7 per cent). Furthermore, more girls from

consistently least-poor households are found to complete elementary education on age than boys, while the situation is reversed among children from chronically poor households. This is indicative of the fact that gender differences exist even among girls based on the wealth

status of households.

Analysis of achievement levels for the Older Cohort reveals that the pattern is similar to that

observed in middle childhood, with relatively small differences by gender, with girls outperforming boys, and large differences seen by place of residence, maternal education

and baseline wealth index. Table 26 presents the results for mathematics and reading.13

Table 26. Achievement in mathematics and reading (Cloze) tests of the Older Cohort at age 15 (2009) Mathematics (%) Cloze (%) Average 13.9 18.3

Gender Boys 13.8 17.8

Girls 14.0 18.8 Gap 0.2 1.0

Place of residence Urban 20.4 28.1 Rural 11.8 14.6 Gap 8.6 13.5

Maternal education No education 12.1 15.6

Primary 16.3 19.8 Middle 18.6 23.1 Secondary and above 18.1 26.5 Gap 6.5 10.9

Baseline wealth index (R1) Top quintile 19.7 27.1 Bottom quintile 12.3 16.9 Gap 7.4 10.2

Caste Scheduled Caste 10.6 15.2 Scheduled Tribe 12.4 19.4 Backward Class 15.3 19.0 Other Castes 14.8 18.9 Gap 2.9 4.2

Dynamic poverty status Consistently least poor 19.7 24.9 – Boys 18.0 22.5 – Girls 21.3 27.2

Chronically poor 12.6 17.5 – Boys 11.1 15.9 – Girls 14.3 19.3 Gap (Least poor – poor) 7.1 7.4

Source: Young Lives, Older Cohort, Round 3, 2009. Note: Scores are presented as percentages to facilitate understanding. The mathematics test has 30 items and the reading (Cloze) test, 24 items.

13 For reading, we used a technique called Cloze, where children have to complete sentences with missing words in a way that

they will be meaningful.


42

Children from consistently least-poor households performed better in mathematics and Cloze

tests at age 15 compared to children from chronically poor households. Interestingly, girls performed better in both mathematics and Cloze tests than boys across both groups, although

the scores of girls in richer households remained higher. This suggests that those girls who are able to transition to secondary education are perhaps likely to do better than boys.

3.4.4 Early adulthood

This section presents data on Young Lives children from the Older Cohort in 2013, when they

were around 19 years old and therefore would be expected to have completed their senior secondary schooling (Grade 12). Table 27 shows that 71.5 per cent of the Older Cohort had

completed secondary education, with only 45.7 per cent completing 12 years of schooling by the expected age. Finishing school on time is more likely for children from Other Castes, children with mothers with higher levels of education, and those from wealthier homes. It is

important to note that by age 19, around 28% of the sample had dropped out of school, the majority of whom were girls, belonged to the poorest quintile, had mothers with low levels of education and lived in rural areas.

Table 27. Completed secondary education by age 19 (Older Cohort) Completed

secondary on age (%)

Completed secondary

overage (%)

Still in secondary (%)

Dropped-out (%)

Undivided Andhra Pradesh 45.7 25.8 0.5 28.0

Gender Boys 49.0 27.1 0.6 23.2 Girls 42.5 24.5 0.4 32.6 Gap 6.5 2.6 0.2 -9.3

Place of residence Urban 55.6 25.8 0.4 18.2 Rural 42.7 25.8 0.6 31.0 Gap 12.9 0.0 -0.1 -12.8

Maternal education No education 38.1 26.5 0.5 34.9 Primary 56.0 28.0 1.0 15.0 Middle 62.2 25.6 1.2 11.0 Secondary and above 76.5 18.4 0.0 5.1 Gap -38.4 9.6 1.2 29.8

Baseline wealth index (R1) Top quintile 63.7 24.7 0.0 11.6 Bottom quintile 30.4 26.2 0.0 43.5

Gap 33.3 -1.4 0.0 -31.9

Caste Scheduled Caste 37.8 28.4 1.0 32.8 Scheduled tribe 41.0 28.6 0.0 30.5 Backward Class 45.6 24.8 0.2 29.4 Other Caste 56.7 23.7 1.0 18.7 Gap -18.9 4.8 -1.0 14.1

Dynamic poverty status Consistently least poor 64.0 24.5 1.0 10.5

– Boys 61.4 25.7 1.0 11.9 – Girls 66.7 23.2 1.0 9.1 Chronically poor 29.0 31.1 0.5 39.5 – Boys 34.3 31.4 1.0 33.3 – Girls 22.7 30.7 0.0 46.6 Gap (Least poor – poor) 35.1 -6.6 0.5 -29.0



43

The largest gap regarding completion of secondary education on age is seen between

children from consistently least-poor households (64 per cent) and those belonging to chronically poor households (29 per cent). There are very large differences in dropout rates

seen among girls and boys belonging to households with different levels of poverty: more girls (46.6 per cent) than boys (33.3 per cent) dropped out of school among chronically poor households, whilst fewer girls (9.1 per cent) than boys (11.9 per cent) dropped out from the

least-poor households.

Dropping out of school becomes a relevant topic for policy and research during late

adolescence. Further analysis of when boys and girls dropped out is necessary to plan targeted preventative interventions. Table 28 shows that the highest incidence of school

dropout occurred by Grade 8 (67.5 per cent), with a steep increase in both boys and girls dropping-out in the grades leading up to this (Grades 6-8). During secondary schooling, a further 32.5 per cent of the children dropped out, something which may be explained by the

Grade 10 Board Exam which many children may not expect to pass. Further gender analysis of school dropout reveals that, 62.8 per cent boys and 70.1 per cent girls dropped out of school before reaching Grade 8 or entering secondary education. Interestingly, fewer girls

dropped out in Grades 9 and 10, compared to boys.

Table 28. Grade in which child dropped out of school

Elementary education Overall (%) Boys (%) Girls (%)

Grade 2 0.9 1.1 0.8

Grade 3 3.6 2.1 4.6

Grade 4 6.8 8.5 5.4

Grade 5 5.9 5.3 6.2

Grade 6 13.5 12.8 13.8

Grade 7 11.7 12.8 10.8

Grade 8 25.2 20.2 28.5

Secondary education

Grade 9 11.3 14.9 9.2

Grade 10 21.2 22.3 20.8

Total 100 100 100

Source: Young Lives study (2013)

It is also important to highlight that, by age 19, around 39.7 per cent of Older Cohort children

had enrolled either in universities or technical or vocational institutes. As shown in Table 29, however, there are differences observed on who went to what type of institution. Table 29 shows that children with more educated mothers (67.3 per cent), those with a high wealth

index (54.7 per cent) and those located in urban areas (46.7 per cent) were more likely to be enrolled in universities. On the other hand, only 7.6 per cent of children were enrolled in vocational courses at age 19, mostly from Scheduled Caste households.


44

Table 29. Type of tertiary education at age 19 (2013)

University (%) Institute (technical or vocational) (%)

Average 32.1 7.6

Gender

Boys 33.7 10.2

Girls 30.5 5.0

Gap 3.2 5.2

Place of residence

Urban 46.7 5.8

Rural 27.5 8.1

Gap 19.2 2.3

Maternal education

No education 23.3 7.6

Primary 44.0 3.0

Middle 47.6 9.8

Secondary and above 67.3 11.2

Gap 44.0 8.2


Top quintile 54.7 7.9

Bottom quintile 16.8 5.2

Gap 37.9 2.7

Caste

Scheduled Caste 20.6 9.3

Scheduled Tribe 30.5 5.7

Backward Class 30.3 7.3

Other Castes 48.3 7.4

Gap 27.7 3.6


Consistently least poor 53.0 10.0

– Boys 50.5 14.9

– Girls 55.6 5.1

Chronically poor 17.9 4.7

– Boys 22.5 4.9

– Girls 12.5 4.5



The relationship between the type of tertiary education attended at age 19 and dynamic

poverty status reveals that 53 and 10 per cent of children from consistently least-poor households are enrolled in universities and technical/vocational institute respectively, compared to only 17.9 and 4.7 per cent of children from chronically poor households. The

gaps in enrolment in universities and vocational institutes between consistently least poor and chronically poor are 35.1 and 5.3 percentage points respectively.

Regarding university enrolment, more girls (55.6 per cent) are found to be enrolled than boys

(50.5 per cent) in consistently least-poor households, although more boys are enrolled in vocational institutes. However, among children from chronically poor households, higher

numbers of boys are enrolled in both universities (22.5 per cent) and vocational institutes (4.9 per cent) than girls (12.5 per cent and 5.3 per cent respectively).


45

Table 30 presents the results of a mathematics and reading tests administered to Young

Lives children at age 19.14 The gap between boys and girls was relatively small in comparison to gaps among other variables. The largest difference in mathematics score at

age 19 was found to be related to maternal education, followed by rich and poor households and between Other Caste and Scheduled Caste children.

Table 30. Achievements in mathematics and reading tests among the Older Cohort at age 19 (2013)

Mathematics (mean scores) Reading (mean scores)

Average 40.5 59.8

Gender

Boys 44.8 62.2

Girls 36.3 57.5

Gap 8.5 4.7

Place of residence

Urban 48.1 64.5

Rural 38.1 58.3

Gap 10.0 6.2

Maternal education

No education 35.7 56.8

Primary 45.7 63.0

Middle 52.9 66.2

Secondary and above 59.3 71.7

Gap 23.6 14.9


Top quintile 51.9 67.0

Bottom quintile 30.4 52.9

Gap 21.5 14.1

Caste

Scheduled Caste 33.9 57.1

Scheduled Tribe 35.8 57.8

Backward Class 40.2 58.8

Other Caste 50.0 65.6

Gap 16.1 8.5


Consistently least poor 53.1 67.6

– Boys 56.3 68.5

– Girls 49.9 66.7

Chronically poor 29.8 53.2

– Boys 34.3 57.2

– Girls 24.5 48.5


Note: Scores are presented as a percentage of correct answers over total possible score. The mathematics test has 29 items; the reading test has 24 items. Source: Young Lives, Older Cohort, Round 4, 2013.

Analysis of these learning achievement data with dynamic poverty status show that children

from consistently least-poor households performed better in both mathematics and reading

tests at age 19 compared to children belonging to chronically poor households. Gender 14 Language test in Telegu.


46

analysis reveals that girls’ performances in both mathematics and reading tests at age 19 are relatively poorer than their male counterparts, irrespective of their poverty status – something which is markedly different to girls’ performance at age 15, where they fared better than

boys.

Many of the children at 19 were engaged in work (35.9 per cent) and were no longer

pursuing their academic studies. Figure 7 shows their occupational situation, with more boys than girls enrolled in tertiary education or the ‘only work/ full time paid work’ category. At 19 ,

37 per cent of the girls were married, with close to 28 per cent getting married before the legal age of eighteen (Singh andVennam,2016)

Interestingly, only a marginal difference (3.7 per cent) was found between male and female

enrolment in universities, although more boys were found to be combining education and

work at tertiary level.

Figure 7. Adolescent´s occupational situation at age 19 (2013)

Source: Young Lives, Older Cohort, Round 4, 2013. Notes: F = female; M = male *In school includes children who were enrolled in Grade 9 to Grade 12 in 2013.

Table 31 shows that, among girls who were studying full time at age 19, less than 7 per cent

were married, whilst 53 per cent of the married girls were working full time. Perhaps more importantly, 66 per cent of girls whose occupation was classified as ‘Other’ (e.g. not full-time education or full-time work or a combination of both) were married, indicating that the

majority of the married girls were engaged in other activities.

Table 31. Marital status and occupation of the Older Cohort female children at age 19 (2013)

Unmarried (%) Married (%) Sample

Studying full-time 93.4 6.6 168

Working full-time 47.4 52.6 156

Working and studying 100.0 0.0 35

Not studying or working 34.2 65.8 123

Total 63.9 36.1 482

Occupational situation

Tertiary education (39.7%) M: 21.7 % F: 18.0 %

Institute (7.6%) M: 4.8% F: 2.4%

Also work (1.5%) M: 1.1% F: 0.4%

University (32.1%) M: 16.6% F: 15.5%

Also work (9.9%) M:7.3% F:2.6%

Only work (35.9%) M: 19.4% F: 16.5%

In school (9.3%) M: 6.3% F: 3.0%

Also work (3.1%) M: 2.6%

F:0.5%

Other (15.1%) M: 1.7% F: 13.4%


47

4. Educational outcomes: gaps over time The key longitudinal findings for both Younger and Older Cohorts at ages 8, 12 and 15 are:

1. A decline in learning achievement across time, despite increase in enrolment

Despite an increase in the enrolment of children at age 12 into elementary school, learning

outcomes (as measured by mathematics test scores) had declined over the same period. This is a matter of policy concern, and suggests that a focus on learning outcomes is necessary to ensure that the poorest children are given an opportunity to learn.

Figure 8. Mathematics scores at age 12

Source: Young Lives, Younger Cohort (Round 4, 2013) and Older Cohort (Round 2, 2006).

2. Many more children are overage in elementary school

While the proportion of overage children seems to be increasing as the Older Cohort children

move from elementary to secondary education (Figure 9), it is surprising that the percentage of overage children was much higher for the Younger Cohort at both age 8 as well as at age 12 than it had been for the earlier cohort. Given the ‘no-detention’ policy instituted as part of

the RTE Act, this is unexpected and extremely difficult to explain. It is potentially related to the increased number of Younger Cohort children attending private schools, which are not closely regulated for implementation of the ‘no detention’ policy, and where children moving

into these schools may be asked to repeat classes.

52.8

65.3

0

10

20

30

40

50

60

70

80

90

100


Perc

enta

ge


48

Figure 9. Overage by age and cohort


3. Gaps in PPVT scores diminishing across caste groups

Figure 10 shows how the gap between the PPVT scores of Scheduled Caste, Scheduled

Tribes and Other Caste Younger Cohort children diminishes over time.

Figure 10. PPVT score by caste (SC, ST, and OC) by age for Younger Cohort

Source: Young Lives, Younger Cohort, R3 to R4 (2009 -2013).

Notes: Scores are presented as a percentage of correct answers over total possible score.

*Ravens scores have been taken for Older Cohort children at age 8.

42.6 46.7

29.7 33.9 36.0

0

10

20

30

40

50

60

70

80

90

100

At age 8 At age 12 At age 15

In P

erce

ntag

e


28.4

75.1

24.7

71.1

33.7

77.8

0

10

20

30

40

50

60

70

80

90

100

At age 8 At age 12

In P

erce

ntag

e

SC ST OC


49

4. PPVT score at age 5 has a strong association with mathematics score at age 8 and 12

The relationship between PPVT scores at age 5 and mathematics scores at later ages for the Younger Cohort is shown in Figure 11. This implies that the receptive vocabulary skills at age

5 may be highly predictive for skills in mathematics, even after seven years. This association has been observed in other studies, and is known as the Mathew Effect; the main idea, as applied to education, is that children with higher skills at an early age will tend to learn at

higher levels than children with lower skills.

Figure 11. Mathematics score at age 8 and 12 by terciles of PPVT score at age 5 (Younger Cohort)

Source: Young Lives, Younger Cohort, R2 to R4 (2006 -2013).

Note: Scores are presented as percentages to facilitate understanding.

5. Achievements and challenges in education This report has presented the main characteristics of the Indian education system, and then

built upon this context by describing findings relating to educational outcomes at different ages by selected background variables using Young Lives longitudinal data from Andhra

Pradesh and Telangana. The advantage of the Young Lives dataset is two-fold: it allows us to understand the evolution of children over time, and to compare indicators across two cohorts of children from the same groups/categories at the same age but at different points in

time. In doing this analysis, our emphasis has been on understanding how educational opportunities and outcomes are linked with the socio-economic and individual characteristics of children. We have focused on gender, caste, place of residence, maternal education, and

examined both static and dynamic poverty. The findings presented have also benefited from the review of other existing literature and studies which complement the scenario of the

0

10

20

30

40

50

60

Bottom Middle Top

PPVT Scores at Age 5

In P

erce

ntag

e

Maths at age 8 Maths at age 12


50

Indian education system and also point towards the challenges ahead. Below are some of the key ideas that come out of the analysis presented in this report, with suggestive policy implications in the Indian context:

1. Enrolment at elementary and secondary level has shown marked improvement

over the past decade. Overall progress is evident in enrolment at all levels of education; however, enrolment is higher at primary level, and becomes progressively lower at upper primary, secondary and tertiary levels, which is an indication of

increasing dropout rates with higher levels of education.

2. Increasing disparity due to enrolment of children from less poor families into

private schools. Enrolment into private schools has doubled for 8 year olds in 2010 as compared to 2002. Children with certain characteristics are found to be more likely

to be enrolled in private schooling, leading to socio-economic and aspirational disparities across the education available to children within different households. For example, boys, children from urban areas, children belong to other castes, and

children who have a mother with a higher level of education are more likely to be enrolled in private fee-charging schools. It is clear from our analysis that children who belong to the consistently least-poor households are more likely to be enrolled in

school (and particularly in private school) as compared to children from chronically poor households between 2002 and 2013.

3. More overage children in recent times. More children are found to be overage

among the Younger Cohort children at age 8 and 12 in 2010 and 2013 than was the case for the Older Cohort children in 2002 and 2006. This is a serious issue, as there

is ‘no detention’ policy in India as an outcome of Right to Education Act (2009), and it may be due to due to the larger number of children changing schools and moving to private schools.

4. Learning achievement has declined over time. Importantly, Younger Cohort children

at age 12 (2013) performed worse in the mathematics achievement test compared to Older Cohort children at the same age (2006). This finding may indicate that, while a larger number of children are accessing school, lack of attention to quality and

learning has resulted in worse learning outcomes for children as the number of children in school increases. Teacher inability to deal with increased diversity in the classroom may well be a cause of this; however, further studies would be required to

test this hypothesis. In addition, we observe huge disparities in achievement scores by gender, caste, place of residence, and both static and dynamic poverty.

5. It is critical to invest in pre-school education, since gaps in achievement are

evident by the age of 5. This report also indicates that attending preschool in early

years is helpful in cognitive achievement at age 5 as measured by PPVT scores. Not only that, children who attended private preschools did better in terms of average PPVT scores, followed by children who attended public preschool. The worst

performers in PPVT scores at age 5 were those children who did not attend any preschool. It is also interesting that children with lower scores at an early age tend to have lower achievement levels later at school. They are also more likely to

discontinue education (Singh and Mukherjee 2016). The effects of preschool in general, and of private preschool in particular, on achievement scores also exacerbates existing disparities, as certain advantaged groups of children such as

boys, urban children, children belonging to Other Castes and least-poor households are more likely to be enrolled in private preschools than their counterparts.


51

66.. EEdduuccaattiioonnaall ooppppoorrttuunniittiieess aanndd iinnvveessttmmeennttss iinn bbaassiicc eedduuccaattiioonn aarree ddiissttrriibbuutteedd iinn aann

uunnffaaiirr wwaayy. The infrastructure facilities of private schools seem to be better than those of public schools, although teachers in public schools were better qualified than

those in private low-fee charging schools. In terms of teaching, many more teachers were observed to check and provide feedback to students in private schools than in public schools, although both public and private schools were using the lecture

method of teaching.

77.. IInnccrreeaassiinngg nnuummbbeerr ooff cchhiillddrreenn ddrroopp oouutt aass tthheeyy ttrraannssiittiioonn ttoo sseeccoonnddaarryy eedduuccaattiioonn. In

our sample we found that around 28 per cent of children had dropped out of secondary school by age 19. Again, the factors linked with dropping out of secondary

schooling are linked largely with households’ level of poverty along with other factors like gender, caste and mothers’ level of education. Child labour at age 12 and writing and reading ability at age 8 emerge as important determinants of secondary school

completion (Singh and Mukherjee, 2016).

8. Access to tertiary education is highly segregated. In India, there are two types of

post-secondary studies: university (colleges/ universities) and technical institutes. Enrolment in tertiary education is determined by many factors, like gender status,

residential status, poverty at households and mother’s education and aspiration.

The above scenarios indicate the progress and challenges within the education system in

India. The low educational achievement for children from economically and socially disadvantaged parts of society needs to be addressed urgently to ensure that education becomes an equaliser. The educational outcomes of children at different ages are mostly

found to be linked with household poverty, along with additional factors like gender, caste and location. This requires consistent policies, programmes and public investment targeting educationally vulnerable children from an early age. We conclude that the educational

system may be reinforcing inequalities since it does not provide equal opportunities for all; ideally, it should prioritise the opportunities of children and young adults who we know are more likely to have poor educational outcomes.

From the ‘life-course’ perspective, the focus should be on early years of education. The link

between early literacy and secondary school completion requires policy attention (Singh and Mukherjee 2015). It is critical that preschool and primary teachers are effectively trained to develop early literacy to ensure children both acquire skills and retain interest in learning.

The ongoing process to formulate a new National Policy on Education is an opportunity which must be seized in this regard.

We also see that, within the Young Lives sample, children in private schools seem to perform

better academically, despite better-qualified and more experienced teachers working in

government schools. In this context it is critical that education policies at macro and micro level serve to promote inclusion, not exclusion, for children from lower castes or poorer areas. It has increasingly become clear that Education For All (EFA) goals cannot be

achieved merely by achieving universal enrolment. Quality education must also become an inherent right for every child irrespective of gender, caste, ethnicity, religion, socio-economic background, ability, or location.

Children in private schools had better mathematics scores than children in government schools, which is perhaps not surprising since the private system caters to more advantaged

children. However, we continue to see this difference in maths scores when we control for children’s previous achievement and several household, child, class and school-level


52

characteristics (Singh and Sarkar 2012). Delivering the quality education required under Education For All commitments and managing the equity implications of a rapidly growing private school system raises some important challenges: (1) raising quality across the board

– including ensuring common quality standards within government and private schools; (2) greater accountability mechanisms, enabling government schools to meet the aspirations and educational needs of the children and their families that they are meant to serve; (3)

ensuring different children are not segregated by sector depending on their location, ethnicity, gender or family wealth, so that education fulfils the potential to become equalising, not reinforcing, of existing divisions.

We see that gender differences become more significant as children get older, but boys are

also not always advantaged – particularly for those belonging to chronically poor households. Policies to reduce gender-based differences are required and engagement with civil society and communities and families is necessary to reduce gender-based discrimination.

For education to become an equaliser, systemic reform is required to safeguard against the

stratification of schools. Regulatory mechanisms must be introduced at the state, district and sub-district levels, to promote school effectiveness. There is a strong case for an autonomous department of standards and evaluation to be set up at the Central and state

level and quality standards developed and implemented across all schools. A ’common school system’ may be the best way forward to ensure that the education system is able to promote equity and social justice. Also, education does not always compensate for

background disadvantage, and may reinforce inequalities. Well-designed school curricula that focus on supporting all children can help to narrow gaps in achievement.

Social protection is a key way of underpinning pro-poor policy, but must be well-designed,

targeted and implemented, and must take into account how programmes are perceived. Benchmarks and regulatory mechanisms need to be developed and implemented for quality

assurance – in both government and private schools. One way to achieve this would be to create an autonomous body for assessment of schools and learning levels of children in every state. This information should be in the public domain to foster greater accountability of

all schools.

On the policy side, Young Lives research highlights serious equity concerns with children

having very different educational opportunities because of their household wealth level, location or gender, with government schools often the only option available to the poorest

households. Polices must take a holistic approach in response to the multidimensional nature and consequences of inequality.


53

References Annual Status of Education Report (2015) ASER 2014, Rural, ASER Centre, New Delhi.

http://www.educationforallinindia.com/ASER_2014_report_pratham.pdf (accessed 1 November 2015)

Cueto, S., J. Leon, G. Guerrero, and I. Munoz, I. (2009). Psychometric characteristics of

cognitive development and achievement instruments in Round 2 of Young Lives. Young Lives Technical Note No. 15. Oxford: Young Lives.

Galab, S., V. Kumar, P. Reddy, R. Sing, and U. Vennam (2011) The impact of growth on

Childhood Poverty in Andhra Pradesh, Round 3 Survey Report, Oxford: Young Lives.

Government of India (2005) National Curriculum Framework 2005, Noida: National Council of Educational Research and Training.

Government of India (2007) The Constitution of India, New Delhi: Ministry of Law and

Justice. http://lawmin.nic.in/coi/coiason29july08.pdf (accessed 15 November 2015).

Government of India (2009) The Right to Free and Compulsory Education Act, Gazette

Extraordinary, New Delhi: Ministry of Law and Justice.

MHRD (2009) Framework of Implementation of Rashtriya Madhyamik Shiksha Abhiyan, New

Delhi: Ministry of Human Resources Development. http://mhrd.gov.in/sites/upload_files/mhrd/files/upload_document/Framework_Final_RMSA_3.

pdf (accessed 2 February 2013)

Planning Commission (2012) Twelfth Five Year Plan (2012–2017), Social Sectors, Vol. III,

New Delhi: Government of India.

NUEPA (2014a) Elementary Education in India: Progress Towards UEE, District Information

System of Education 2013–14, New Delhi: National University of Educational Planning and Administration (NUEPA).

NUEPA (2014b) Education For All Towards Quality with Equity, India, New Delhi: National

University of Educational Planning and Administration (NUEPA).

NUEPA (2014 c) Secondary Education in India: Progress Towards Universalisation, Unified

District Information System of Education 2013–14, New Delhi: National University of Educational Planning and Administration (NUEPA).

RMSA (2014) Third Joint Review Mission: 13–27 January 2014. Aide Memoire, New Delhi: Ministry of Human Resource Development.

http://mhrd.gov.in/sites/upload_files/mhrd/files/upload_document/3rd%20Jrm%20Aide%20Memmiore.pdf (accessed 4 September 2014)

Sarva Shiksha Abhiyan (2000) Ministry of Human Resource Development, Government of

India.

Singh, R. and C. Bangay (2014) ‘Low Fee Private Schooling in India – More Questions than

Answers? Observations from the Young Lives Longitudinal Research in Andhra Pradesh’, International Journal of Educational Development 39: 142–50.

Singh, R. and P. Mukherjee (2015) Determinants of Successful Completion of Secondary

Education: Evidence from Andhra Pradesh, Young Lives Working Paper 143, Oxford: Young Lives.


54

Singh, R. and P. Mukherjee (2016) ‘Comparison of Public & Private Pre-school Education on

Developmental Outcomes of Children: Evidence from Young Lives, India’, Young Lives Working Paper 143, Oxford: Young Lives (forthcoming).

Singh, R. and S. Sarkar (2012) Teaching Quality Counts: How Student Outcomes Relate to

Quality of Teaching in Private and Public Schools in India, Young Lives Working Paper 91, Oxford: Young Lives.

SRI-IMRB (2009) ‘National Sample Survey Estimation of Out-of-School Children of age 6-13

years in India’, New Delhi. Study commissioned by the MHRD&Ed.Cil, New Delhi.

http://ssa.nic.in/pabminutes-documents/NS.pdf (accessed 16 November 2015).

SRI-IMRB (2009) ‘All India Survey of Out-of-School Children of age 6-13 years and age 5’,

New Delhi. Study commissioned by the MHRD&Ed.Cil, New Delhi.

SRI-IMRB (2005) ‘All India Survey of Out-of-School Children in the 6-13 Years Age Group’,

Study commissioned by the MHRD&Ed.Cil, New Delhi. http://www.educationforallinindia.com/Survey-Reportof-20out-of-school-children-IMRB-

MHRD-EDCil-2009.pdf (accessed 12 May 2011).

UNESCO-UIS (2015) A growing number of children and adolescents are out of school as aid

fails to meet the mark, Policy paper 22/Fact sheet 31, UNESCO-UIS. http://www.uis.unesco.org/Education/Documents/fs-31-out-of-school-children-en.pdf

(accessed 1 Sept 2015).

Woodhead, M., P. Ames, U. Vennam, W. Abebe, and N. Streuli (2009) Equity and quality?

Challenges for early childhood and primary education in Ethiopia, India and Peru, Working Paper 55,The Hague: Bernard van Leer Foundation.

Woodhead, M., M. Frost, and Z. James (2011) ‘Does growth in private schooling contribute to

Education For All? Evidence from a longitudinal two cohort study in Andhra Pradesh, India.’

Draft article prepared for International Journal of Educational Development.


55

Appendices

Appendix A

Table A1. Profiling of castes from Young Lives (all in percentages)

Caste Place of residence Mother's education Mean Wealth Index

Urban Rural No formal education

Secondary and above education

Baseline Wealth

Index (2002)

Current Wealth Index

(2013)

Older Cohort

Scheduled Caste 13.7 86.3 88.2 2.9 0.33 0.54

Scheduled Tribe 8.3 91.7 79.3 8.5 0.32 0.52

Backward Class 24.4 75.6 70.8 9.7 0.41 0.62

Other Castes 45.2 54.8 42.6 23.2 0.53 0.69

Younger Cohort

Scheduled Caste 15.1 84.9 69.8 15.2 0.36 0.53

Scheduled Tribe 6.5 93.5 77.7 8.6 0.25 0.47

Backward Class 23.2 76.8 62.6 20.3 0.41 0.60

Other Castes 50.9 49.1 29.3 45.4 0.55 0.69

Appendix B

Table B1. Enrolment in preschool by type of school

Older Cohort Younger Cohort

Undivided Andhra Pradesh Private 31.6 37.7

Public 68.4 62.3

Urban Private 80.8 87.8

Public 19.2 12.2

Rural Private 13.8 17.7

Public 86.2 82.3


56

Appendix C

Figure C1. Average mathematics score by private and public school among Younger Cohort children

Table C1. Average mathematics score by gender among Younger Cohort children

At age 8 (2009) At age 12 (2013)

Boys 41.9 44.0

Girls 41.0 44.0

Figure C2. Average mathematics score by place of residence among Younger Cohort children

45.4

53.7

38.6 40.4

0

10

20

30

40

50

60

At age 8 (2009) At age 12 (2013)

Aver

age

Scor

e in

Mat

hem

atic

s

Private Public

0

10

20

30

40

50

60

At age 8 (2009) At age 12 (2013)

Aver

age

Scor

e in

Mat

hem

atic

s

Urban Rural


57

Figure C3. Learning outcomes of 12-year-old Older and Younger Cohort children in public and private schools

Appendix D

Figure D1. Association between mathematics scores and wealth index per pupil at age 12, by clusters, Younger Cohort

Note: Pearson correlation 0.43

0

10

20

30

40

50

60

70

80

90

100

"Which of these is equal to 342?"

"Which of these is the name for

9740?"

"A piece of rope 204 cm. long is cut into 4 equal pieces.

Which of these gives the length of

each piece in centimeters?"

Average score of 3 Maths Qs

% o

f YL

Chi

ldre

n w

ith C

orre

ct A

nsw

ers

Public

2006 2013

0

10

20

30

40

50

60

70

80

90

100

"Which of these is equal to 342?"

"Which of these is the name for

9740?"

"A piece of rope 204 cm. long is cut into 4 equal pieces.

Which of these gives the length of

each piece in centimeters?"

Average score of 3 Maths Qs

% o

f YL

Chi

ldre

n w

ith C

orre

ct A

nsw

ers

Private

2006 2013

0

10

20

30

40

50

60

70

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

Mat

hem

atic

s Sc

ore

In P

erce

ntag

e

Baselevel Wealth Index (2002)

Maths score at age 12 Linear (Maths score at age 12) Linear (Maths score at age 12)


58

Appendix E

Table E1. Attendance rate at preschool at age 5 (Older Cohort)

Older Cohort Public school Private school Not attending school

Average 41.0 18.9 40.1

Gender

Boys 35.6 20.0 44.4

Girls 46.3 17.8 35.9

Gap 10.7 2.2 8.5

Place of residence

Urban 12.9 54.2 32.9

Rural 49.7 8.0 42.3

Gap 36.8 46.2 9.4

Maternal education

No education 46.6 8.3 45.1

Primary 36.0 28.0 36.0

Middle 34.2 39.0 26.8

Secondary and above 14.3 66.3 19.4

Gap 32.3 58.0 25.7


Top quintile 10.0 60.0 30.0

Bottom quintile 55.5 1.6 42.9

Gap 45.5 58.4 12.9

Caste

Scheduled Caste 48.0 6.4 45.6

Scheduled Tribe 57.1 11.4 31.4

Backward Class 41.5 16.0 42.6

Other Caste 24.6 41.9 33.5

Gap 32.5 35.5 12.1



– Boys 11.9 59.4 28.7

– Girls 21.2 55.6 23.2

Chronically poor 54.2 2.6 43.2

– Boys 52.0 2.9 45.1

– Girls 56.8 2.3 40.9

Gap (Least poor – poor) -37.7 54.9 -17.2


59

Appendix F

Table F1. PPVT score over time (percentages)

Younger Cohort at

age 5 (2006)

Younger Cohort at

age 8 (2009)

Younger Cohort at

age 12 (2013)


(2006)


(2009)

Average 22.0 28.7 75.2 72.2 74.9

Gender

Boys 22.2 30.0 75.4 73.2 74.7

Girls 21.7 27.1 74.9 71.2 75.1

Gap 0.5 2.9 0.6 2.0 0.4

Place of residence

Urban 29.0 33.2 77.4 79.9 87.4

Rural 19.8 27.3 74.5 69.8 70.9

Gap 9.2 6.0 3.0 10.1 16.5

Maternal education

No education 17.7 25.9 72.3 68.6 72.4

Primary 23.7 29.5 76.5 76.0 77.9

Middle 24.4 31.9 78.7 79.7 79.8

Secondary and above 31.5 34.2 80.8 87.5 81.8

Gap 13.9 8.3 8.5 18.9 9.3


Top quintile 32.7 35.1 79.5 81.2 85.9

Bottom quintile 18.2 23.2 69.8 66.6 72.0

Gap 14.5 11.9 9.7 14.5 13.9

Caste

Scheduled Caste 18.0 28.4 75.1 66.9 71.8

Scheduled Tribe 27.4 24.7 71.1 79.8 75.1

Backward Class 19.9 27.9 75.4 70.6 74.7

Other Castes 26.5 33.7 77.8 76.8 78.2

Gap 9.3 9.0 6.8 12.9 6.5


Consistently least poor 31.1 34.7 80.2 81.7 83.0

– Boys 31.0 36.1 79.5 81.0 82.1

– Girls 31.4 32.8 81.1 82.4 83.9

Chronically poor 18.7 22.7 68.7 66.5 71.6

– Boys 19.5 23.2 69.0 66.7 71.4

– Girls 17.8 22.1 68.4 66.3 71.9

Gap (Least poor – poor) 12.4 12.0 11.5 15.1 11.4

Note: Scores are presented as percentages to facilitate understanding. The test has 125 items.


60

Appendix G. Timeline of key educational initiatives and policies

1986 National Policy on Education 1986 (NPE 1986) adopted.

1987 Several large centrally-assisted schemes/programmes such as ‘Operation Blackboard’ and the scheme for restructuring and reorganisation of teacher education launched.

1988 National Literacy Mission (NLM) launched

1992 National Policy on Education 1986 revised.

1994 District Primary Education Programme (DPEP) launched to universalise primary education in selected districts.

1995 Centrally-assisted National Programme of Nutritional Support to Primary Education, popularly known as the Mid-Day Meal Scheme (MDMS) launched.

1999 A separate Department of School Education and Literacy created within the Ministry of Human Resource Development, Government of India.

2001 Sarva Shiksha Abhiyan, the flagship programme for universalisation of elementary education, launched.

2002 The Constitution (Eighty-sixth Amendment) Act, 2002: (i) inserted Article 21-A into the Constitution of India to provide free and compulsory education for all children aged six to 14 years as a fundamental right; (ii) substituted Article 45, which now stated “The State shall endeavour to provide early childhood care and education for all children until they complete the age of six years”.

2003 National Youth Policy, 2003 formulated.

2004 Education Cess introduced for raising additional financial resources needed to fulfil Government’s commitment to universalise elementary education.

2005 National Curriculum Framework (NCF-2005) for school education formulated.

2007 Eleventh Five-Year Plan (2007-12) launched.

2009 (i) The Right of Children to Free and Compulsory Education Act, 2009 enacted;

(ii) The revised National Curriculum Framework for teacher Education formulated; (iii) The Rashtriya Madhyamik Shiksha Abhiyan (RMSA) launched in March 2009, with the vision of making secondary education of good quality available, accessible and affordable to all young persons in the age group 15-16 years; (iv) Revised Centrally-sponsored Scheme of Inclusive Education for the Disabled at Secondary Stage approved; (v) The Centrally-Sponsored Scheme “Construction and Running of Girls’ Hostel for Students of Secondary and Higher Secondary Schools approved.

2010 (i) The Right of Children to Free and Compulsory Education (RTE) Act 2009 came into force from 1 April 2010; (ii) All states/union territories notified of State RTE Rules. Central RTE rules apply to union territories without legislation.

2011 The revised Centrally-Sponsored Scheme “Vocationalisation of Higher Secondary Education” approved.

2012 The Twelfth Five-Year Plan (2007-2012) launched.

2013 (i) National Early Childhood Care and Education (ECCE) Policy adopted; (ii) The Integrated Child Development Services, the flagship programme of Government of India for ECCE restructured and strengthened.

2014 National Youth Policy, 2014 adopted.

Education Trajectories: From Early Childhood to Early Adulthood in India

This report draws upon Young Lives longitudinal data gathered in Andhra Pradesh and Telangana to trace the educational trajectories of two cohorts of children since 2002. From this data, it is clear that huge disparities exist in educational outcomes for children, based on wealth index, gender, location and dynamic poverty status. Stratification of better-off children and boys into private low-fee charging schools adds further to the inequity.

This report highlights the increasing inequities and lack of opportunities afforded to children, particularly girls, from chronically poor households. The low educational achievement for children from economically and socially disadvantaged backgrounds needs to be addressed urgently to ensure that education becomes an equaliser rather than a source of increasing division. This would need consistent policies, programmes and public investment targeting educationally vulnerable children from an early age.

Some policy recommendations are:

• It is critical to focus on early years and ensure that preschool and primary teachers are effectively trained to develop early literacy, so that children acquire skills and retain interest in learning.

• Given the increase in private school enrolment of children from better-off households, it is important that the Sustainable Development Goals’ focus on ‘quality education’ must become an inherent right for every child irrespective of gender, caste, ethnicity, religion, socio-economic background, ability, or location.

• Since gender disparity in schooling increases as children grow older, safety nets and social protection for the poorest households must be made available.

• There is a strong case for an autonomous department of standards and evaluation to be set up and quality standards developed and implemented across all schools. A ‘common school system’ may be the best way forward to ensure that the education system is able to promote equity and social justice.

About Young Lives

Young Lives is an international study of childhood poverty, involving 12,000 children in 4 countries over 15 years. It is led by a team in the Department of International Development at the University of Oxford in association with research and policy partners in the 4 study countries: Ethiopia, India, Peru and Vietnam.

Through researching different aspects of children’s lives, we seek to improve policies and programmes for children.

Young Lives Partners

Young Lives is coordinated by a small team based at the University of Oxford, led by Professor Jo Boyden.

• Ethiopian Development Research Institute, Ethiopia

• Pankhurst Development Research and Consulting plc, Ethiopia

• Centre for Economic and Social Studies, Hyderabad, India

• Save the Children India

• Sri Padmavathi Mahila Visvavidyalayam (Women’s University), Andhra Pradesh, India

• Grupo de Análisis para el Desarollo (GRADE), Peru

• Instituto de Investigación Nutricional, Peru

• Centre for Analysis and Forecasting, Vietnamese Academy of Social Sciences, Vietnam

• General Statistics Office, Vietnam

• Oxford Department of International Development, University of Oxford, UK

Contact:Young LivesOxford Department of International Development,University of Oxford,3 Mansfield Road,Oxford OX1 3TB, UKTel: +44 (0)1865 281751Email: [email protected]: www.younglives.org.uk

www.younglives.org.uk